Daily Papers

Transformers yield state-of-the-art results across many tasks. However, their heuristically designed architecture impose huge computational costs during inference. This work aims on challenging the common design philosophy of the Vision Transformer (ViT) model with uniform dimension across all the stacked blocks in a model stage, where we redistribute the parameters both across transformer blocks and between different structures within the block via the first systematic attempt on global structural pruning. Dealing with diverse ViT structural components, we derive a novel Hessian-based structural pruning criteria comparable across all layers and structures, with latency-aware regularization for direct latency reduction. Performing iterative pruning on the DeiT-Base model leads to a new architecture family called NViT (Novel ViT), with a novel parameter redistribution that utilizes parameters more efficiently. On ImageNet-1K, NViT-Base achieves a 2.6x FLOPs reduction, 5.1x parameter reduction, and 1.9x run-time speedup over the DeiT-Base model in a near lossless manner. Smaller NViT variants achieve more than 1% accuracy gain at the same throughput of the DeiT Small/Tiny variants, as well as a lossless 3.3x parameter reduction over the SWIN-Small model. These results outperform prior art by a large margin. Further analysis is provided on the parameter redistribution insight of NViT, where we show the high prunability of ViT models, distinct sensitivity within ViT block, and unique parameter distribution trend across stacked ViT blocks. Our insights provide viability for a simple yet effective parameter redistribution rule towards more efficient ViTs for off-the-shelf performance boost.

Neural architecture search (NAS) has allowed for the automatic creation of new and effective neural network architectures, offering an alternative to the laborious process of manually designing complex architectures. However, traditional NAS algorithms are slow and require immense amounts of computing power. Recent research has investigated training-free NAS metrics for image classification architectures, drastically speeding up search algorithms. In this paper, we investigate training-free NAS metrics for recurrent neural network (RNN) and BERT-based transformer architectures, targeted towards language modeling tasks. First, we develop a new training-free metric, named hidden covariance, that predicts the trained performance of an RNN architecture and significantly outperforms existing training-free metrics. We experimentally evaluate the effectiveness of the hidden covariance metric on the NAS-Bench-NLP benchmark. Second, we find that the current search space paradigm for transformer architectures is not optimized for training-free neural architecture search. Instead, a simple qualitative analysis can effectively shrink the search space to the best performing architectures. This conclusion is based on our investigation of existing training-free metrics and new metrics developed from recent transformer pruning literature, evaluated on our own benchmark of trained BERT architectures. Ultimately, our analysis shows that the architecture search space and the training-free metric must be developed together in order to achieve effective results.

Local motion blur commonly occurs in real-world photography due to the mixing between moving objects and stationary backgrounds during exposure. Existing image deblurring methods predominantly focus on global deblurring, inadvertently affecting the sharpness of backgrounds in locally blurred images and wasting unnecessary computation on sharp pixels, especially for high-resolution images. This paper aims to adaptively and efficiently restore high-resolution locally blurred images. We propose a local motion deblurring vision Transformer (LMD-ViT) built on adaptive window pruning Transformer blocks (AdaWPT). To focus deblurring on local regions and reduce computation, AdaWPT prunes unnecessary windows, only allowing the active windows to be involved in the deblurring processes. The pruning operation relies on the blurriness confidence predicted by a confidence predictor that is trained end-to-end using a reconstruction loss with Gumbel-Softmax re-parameterization and a pruning loss guided by annotated blur masks. Our method removes local motion blur effectively without distorting sharp regions, demonstrated by its exceptional perceptual and quantitative improvements compared to state-of-the-art methods. In addition, our approach substantially reduces FLOPs by 66% and achieves more than a twofold increase in inference speed compared to Transformer-based deblurring methods. We will make our code and annotated blur masks publicly available.

Recently, large language models (LLM) based on transformers are facing memory bottleneck issues due to KV cache, especially in long sequence handling. Previous researches proposed KV cache compression techniques that identify insignificant tokens based on Accumulative Attention Scores and removes their items from KV cache, noting that only few tokens play an important role in attention operations. However, we have observed that the existing Accumulative Attention Score is not suitable for the transformer decoder structure. In the decoder model, the number of times the Attention Score accumulates varies depending on the order of token appearance due to the effect of masking, causing an uneven comparison between tokens. To solve this, we propose Accumulative Attention Score with Forgetting Factor (A2SF) technique, which introduces a Forgetting Factor in the Attention Score accumulation process. A2SF applies a penalty to the past Attention Score generated from old tokens by repeatedly multiplying the Forgetting Factor to the Attention Score over time. Therefore, older tokens receive a larger penalty, providing fairness among different ages of tokens. Through the fair comparison among tokens, we can more effectively select important tokens. We have verified the accuracy improvement through A2SF in the OPT and LLaMA models and A2SF improves the accuracy of LLaMA 2 by up to 7.8% and 5.1% on 1-shot and 0-shot.

Transformer models have revolutionized natural language processing with their unparalleled ability to grasp complex contextual relationships. However, the vast number of parameters in these models has raised concerns regarding computational efficiency, environmental impact, and deployability on resource-limited platforms. To address these challenges, this paper investigates the application of weight pruning-a strategic reduction of model parameters based on their significance-as an optimization strategy for Transformer architectures. Through extensive experimentation, we explore various pruning methodologies, highlighting their impact on model performance, size, and computational demands. Our findings suggest that with judicious selection of pruning hyperparameters, significant reductions in model size are attainable without considerable compromise on performance. Moreover, when coupled with post-pruning fine-tuning strategies, some pruned models even exhibit enhanced generalization capabilities. This work seeks to bridge the gap between model efficiency and performance, paving the way for more scalable and environmentally responsible deep learning applications.

Multilingual transformer-based models demonstrate remarkable zero and few-shot transfer across languages by learning and reusing language-agnostic features. However, as a fixed-size model acquires more languages, its performance across all languages degrades, a phenomenon termed interference. Often attributed to limited model capacity, interference is commonly addressed by adding additional parameters despite evidence that transformer-based models are overparameterized. In this work, we show that it is possible to reduce interference by instead identifying and pruning language-specific parameters. First, we use Shapley Values, a credit allocation metric from coalitional game theory, to identify attention heads that introduce interference. Then, we show that removing identified attention heads from a fixed model improves performance for a target language on both sentence classification and structural prediction, seeing gains as large as 24.7\%. Finally, we provide insights on language-agnostic and language-specific attention heads using attention visualization.

Large language models (LLMs) have proven to be highly effective across various natural language processing tasks. However, their large number of parameters poses significant challenges for practical deployment. Pruning, a technique aimed at reducing the size and complexity of LLMs, offers a potential solution by removing redundant components from the network. Despite the promise of pruning, existing methods often struggle to achieve substantial end-to-end LLM inference speedup. In this paper, we introduce SLEB, a novel approach designed to streamline LLMs by eliminating redundant transformer blocks. We choose the transformer block as the fundamental unit for pruning, because LLMs exhibit block-level redundancy with high similarity between the outputs of neighboring blocks. This choice allows us to effectively enhance the processing speed of LLMs. Our experimental results demonstrate that SLEB successfully accelerates LLM inference without compromising the linguistic capabilities of these models, making it a promising technique for optimizing the efficiency of LLMs. The code is available at: https://github.com/leapingjagg-dev/SLEB

Overparametrized transformer networks are the state-of-the-art architecture for Large Language Models (LLMs). However, such models contain billions of parameters making large compute a necessity, while raising environmental concerns. To address these issues, we propose FinerCut, a new form of fine-grained layer pruning, which in contrast to prior work at the transformer block level, considers all self-attention and feed-forward network (FFN) layers within blocks as individual pruning candidates. FinerCut prunes layers whose removal causes minimal alternation to the model's output -- contributing to a new, lean, interpretable, and task-agnostic pruning method. Tested across 9 benchmarks, our approach retains 90% performance of Llama3-8B with 25% layers removed, and 95% performance of Llama3-70B with 30% layers removed, all without fine-tuning or post-pruning reconstruction. Strikingly, we observe intriguing results with FinerCut: 42% (34 out of 80) of the self-attention layers in Llama3-70B can be removed while preserving 99% of its performance -- without additional fine-tuning after removal. Moreover, FinerCut provides a tool to inspect the types and locations of pruned layers, allowing to observe interesting pruning behaviors. For instance, we observe a preference for pruning self-attention layers, often at deeper consecutive decoder layers. We hope our insights inspire future efficient LLM architecture designs.

Pruning is an effective method to reduce the memory footprint and FLOPs associated with neural network models. However, existing structured-pruning methods often result in significant accuracy degradation for moderate pruning levels. To address this problem, we introduce a new Hessian Aware Pruning (HAP) method coupled with a Neural Implant approach that uses second-order sensitivity as a metric for structured pruning. The basic idea is to prune insensitive components and to use a Neural Implant for moderately sensitive components, instead of completely pruning them. For the latter approach, the moderately sensitive components are replaced with with a low rank implant that is smaller and less computationally expensive than the original component. We use the relative Hessian trace to measure sensitivity, as opposed to the magnitude based sensitivity metric commonly used in the literature. We test HAP for both computer vision tasks and natural language tasks, and we achieve new state-of-the-art results. Specifically, HAP achieves less than 0.1%/0.5% degradation on PreResNet29/ResNet50 (CIFAR-10/ImageNet) with more than 70\%/50\% of parameters pruned. Meanwhile, HAP also achieves significantly better performance (up to 0.8\% with 60\% of parameters pruned) as compared to gradient based method for head pruning on transformer-based models. The framework has been open sourced and available online.

We present Statistical Calibrated Activation Pruning (SCAP), a post-training activation pruning framework that (1) generalizes sparsification by input activations of Fully-Connected layers for generic and flexible application across Transformers, and (2) features a simple Mode-Centering technique to pre-calibrate activation distributions for maximizing post-training sparsity. Our results demonstrate robust Pareto efficiency compared to prior methods, translating to a 1.5x additional LLM decoding speedup against CATS at iso model quality. SCAP effectiveness is empirically verified across a wide range of models, including recent Transformer Decoders, MoE, Mamba2, Encoding Transformer, and pre-quantized models, highlighting its practicality and scalability. The code is available at: https://github.com/IntelLabs/SCAP.

We present STAT: a simple algorithm to prune transformer models without any fine-tuning. STAT eliminates both attention heads and neurons from the network, while preserving accuracy by calculating a correction to the weights of the next layer. Each layer block in the network is compressed using a series of principled matrix factorizations that preserve the network structure. Our entire algorithm takes minutes to compress BERT, and less than three hours to compress models with 7B parameters using a single GPU. Using only several hundred data examples, STAT preserves the output of the network and improves upon existing gradient-free pruning methods. It is even competitive with methods that include significant fine-tuning. We demonstrate our method on both encoder and decoder architectures, including BERT, DistilBERT, and Llama-2 using benchmarks such as GLUE, Squad, WikiText2.

Recently vision transformer models have become prominent models for a range of tasks. These models, however, usually suffer from intensive computational costs and heavy memory requirements, making them impractical for deployment on edge platforms. Recent studies have proposed to prune transformers in an unexplainable manner, which overlook the relationship between internal units of the model and the target class, thereby leading to inferior performance. To alleviate this problem, we propose a novel explainable pruning framework dubbed X-Pruner, which is designed by considering the explainability of the pruning criterion. Specifically, to measure each prunable unit's contribution to predicting each target class, a novel explainability-aware mask is proposed and learned in an end-to-end manner. Then, to preserve the most informative units and learn the layer-wise pruning rate, we adaptively search the layer-wise threshold that differentiates between unpruned and pruned units based on their explainability-aware mask values. To verify and evaluate our method, we apply the X-Pruner on representative transformer models including the DeiT and Swin Transformer. Comprehensive simulation results demonstrate that the proposed X-Pruner outperforms the state-of-the-art black-box methods with significantly reduced computational costs and slight performance degradation.

High runtime memory and high latency puts significant constraint on Vision Transformer training and inference, especially on edge devices. Token pruning reduces the number of input tokens to the ViT based on importance criteria of each token. We present a Background Aware Vision Transformer (BAViT) model, a pre-processing block to object detection models like DETR/YOLOS aimed to reduce runtime memory and increase throughput by using a novel approach to identify background tokens in the image. The background tokens can be pruned completely or partially before feeding to a ViT based object detector. We use the semantic information provided by segmentation map and/or bounding box annotation to train a few layers of ViT to classify tokens to either foreground or background. Using 2 layers and 10 layers of BAViT, background and foreground tokens can be separated with 75% and 88% accuracy on VOC dataset and 71% and 80% accuracy on COCO dataset respectively. We show a 2 layer BAViT-small model as pre-processor to YOLOS can increase the throughput by 30% - 40% with a mAP drop of 3% without any sparse fine-tuning and 2% with sparse fine-tuning. Our approach is specifically targeted for Edge AI use cases.

In recent years, there has been a growing emphasis on compressing large pre-trained transformer models for resource-constrained devices. However, traditional pruning methods often leave the embedding layer untouched, leading to model over-parameterization. Additionally, they require extensive compression time with large datasets to maintain performance in pruned models. To address these challenges, we propose VTrans, an iterative pruning framework guided by the Variational Information Bottleneck (VIB) principle. Our method compresses all structural components, including embeddings, attention heads, and layers using VIB-trained masks. This approach retains only essential weights in each layer, ensuring compliance with specified model size or computational constraints. Notably, our method achieves upto 70% more compression than prior state-of-the-art approaches, both task-agnostic and task-specific. We further propose faster variants of our method: Fast-VTrans utilizing only 3% of the data and Faster-VTrans, a time efficient alternative that involves exclusive finetuning of VIB masks, accelerating compression by upto 25 times with minimal performance loss compared to previous methods. Extensive experiments on BERT, ROBERTa, and GPT-2 models substantiate the efficacy of our method. Moreover, our method demonstrates scalability in compressing large models such as LLaMA-2-7B, achieving superior performance compared to previous pruning methods. Additionally, we use attention-based probing to qualitatively assess model redundancy and interpret the efficiency of our approach. Notably, our method considers heads with high attention to special and current tokens in un-pruned model as foremost candidates for pruning while retained heads are observed to attend more to task-critical keywords.

Deploying pre-trained transformer models like BERT on downstream tasks in resource-constrained scenarios is challenging due to their high inference cost, which grows rapidly with input sequence length. In this work, we propose a constraint-aware and ranking-distilled token pruning method ToP, which selectively removes unnecessary tokens as input sequence passes through layers, allowing the model to improve online inference speed while preserving accuracy. ToP overcomes the limitation of inaccurate token importance ranking in the conventional self-attention mechanism through a ranking-distilled token distillation technique, which distills effective token rankings from the final layer of unpruned models to early layers of pruned models. Then, ToP introduces a coarse-to-fine pruning approach that automatically selects the optimal subset of transformer layers and optimizes token pruning decisions within these layers through improved L_0 regularization. Extensive experiments on GLUE benchmark and SQuAD tasks demonstrate that ToP outperforms state-of-the-art token pruning and model compression methods with improved accuracy and speedups. ToP reduces the average FLOPs of BERT by 8.1x while achieving competitive accuracy on GLUE, and provides a real latency speedup of up to 7.4x on an Intel CPU.

The recently proposed Forgetting Transformer (FoX) incorporates a forget gate into softmax attention and has shown consistently better or on-par performance compared to the standard RoPE-based Transformer. Notably, many attention heads in FoX tend to forget quickly, causing their output at each timestep to rely primarily on the local context. Based on this observation, we propose Adaptive Computation Pruning (ACP) for FoX, a method that dynamically prunes computations involving input-output dependencies that are strongly decayed by the forget gate. This is achieved using a dynamically set pruning threshold that ensures that the pruned attention weights remain negligible. We apply ACP to language model pretraining with FoX and show it consistently reduces the number of FLOPs in softmax attention by around 70% across different model sizes and context lengths, resulting in a roughly 10% to 35% improvement in training throughput. Furthermore, longer context lengths yield greater computational savings. All these speed improvements are achieved without any performance degradation. We also perform several analyses to provide deeper insights into our method, such as examining the pruning patterns and analyzing the distribution of FLOP savings across different attention heads. Our code is available at https://github.com/zhixuan-lin/arctic-fox.

Transformer-based Large Language Models (LLMs) have demonstrated remarkable success across various challenging tasks. However, the deployment of LLMs is hindered by their substantial parameter count and memory consumption. Recently, numerous studies have attempted to compress LLMs by pruning them using training-free methods. However, these pruned models often experience significant performance degradation on complex tasks. To address this issue, we propose a novel training pipeline for semi-structured sparse models, named Adaptive Sparse Trainer (AST). By distilling the knowledge stored in its dense counterpart, we prevent the sparse model from overfitting and ensure a stable training process. Moreover, AST allows the model to adaptively select better lottery tickets (e.g., masks) during training. Additionally, we discovered that adding extra well-initialized parameters can further enhance model performance with only a small increase in memory footprint. Our method significantly narrows the performance gap between dense and sparse models while maintaining limited computational cost. Furthermore, when combined with existing quantization methods, AST can compress language models by up to 16x compared to dense FP32 precision models with minimal performance loss. AST outperforms previous state-of-the-art methods by reducing the zero-shot accuracy gap between dense and semi-structured sparse models to 1.12% across multiple zero-shot tasks on Llama2-7B, using less than 0.4% of the pretraining tokens.

Deploying transformer models in practice is challenging due to their inference cost, which scales quadratically with input sequence length. To address this, we present a novel Learned Token Pruning (LTP) method which adaptively removes unimportant tokens as an input sequence passes through transformer layers. In particular, LTP prunes tokens with an attention score below a threshold value which is learned for each layer during training. Our threshold-based method allows the length of the pruned sequence to vary adaptively based on the input sequence, and avoids algorithmically expensive operations such as top-k token selection. We extensively test the performance of LTP on GLUE tasks and show that our method outperforms the prior state-of-the-art token pruning methods by up to ~2.5% higher accuracy with the same amount of FLOPs. In particular, LTP achieves up to 2.1x FLOPs reduction with less than 1% accuracy drop, which results in up to 1.9x and 2.0x throughput improvement on Intel Haswell CPUs and NVIDIA V100 GPUs, respectively. Furthermore, we demonstrate that LTP is more robust than prior methods to variations on input sentence lengths. Our code has been developed in PyTorch and has been open-sourced.

Transformer encoders are widely deployed in large-scale web services for natural language understanding tasks such as text classification, semantic retrieval, and content ranking. However, their high inference latency and memory consumption pose significant challenges for real-time serving and scalability. These limitations stem largely from architectural redundancy, particularly in the attention module. The inherent parameter redundancy of the attention mechanism, coupled with the fact that its attention heads operate with a degree of independence, makes it particularly amenable to structured model compression. In this paper, we propose SHRP (Specialized Head Routing and Pruning), a novel structured pruning framework that automatically identifies and removes redundant attention heads while preserving most of the model's accuracy and compatibility. SHRP introduces Expert Attention, a modular design that treats each attention head as an independent expert, followed by a lightweight shared expander feed-forward network that refines their outputs. The framework employs a unified Top-1 usage-driven mechanism to jointly perform dynamic routing during training and deterministic pruning at deployment. Experimental results on the GLUE benchmark using a BERT-base encoder show that SHRP achieves 93% of the original model accuracy while reducing parameters by 48 percent. Under an extreme compression scenario where 11/12 of the layers are pruned, the model still maintains 84% accuracy and delivers a 4.2x throughput gain while reducing computation to as low as 11.5 percent of the original FLOPs, demonstrating its practical utility for large-scale and latency-sensitive web deployments.

Large language models (LLMs) show excellent performance in difficult tasks, but they often require massive memories and computational resources. How to reduce the parameter scale of LLMs has become research hotspots. In this study, we make an important observation that the multi-head self-attention (MHA) sub-layer of Transformer exhibits noticeable low-rank structure, while the feed-forward network (FFN) sub-layer does not. With this regard, we design a mixed compression model, which organically combines Low-Rank matrix approximation And structured Pruning (LoRAP). For the MHA sub-layer, we propose an input activation weighted singular value decomposition method to strengthen the low-rank characteristic. Furthermore, we discover that the weight matrices in MHA sub-layer have different low-rank degrees. Thus, a novel parameter allocation scheme according to the discrepancy of low-rank degrees is devised. For the FFN sub-layer, we propose a gradient-free structured channel pruning method. During the pruning, we get an interesting finding that the least important 1% of parameter actually play a vital role in model performance. Extensive evaluations on zero-shot perplexity and zero-shot task classification indicate that our proposal is superior to previous structured compression rivals under multiple compression ratios.

Transformer-based NLP models are trained using hundreds of millions or even billions of parameters, limiting their applicability in computationally constrained environments. While the number of parameters generally correlates with performance, it is not clear whether the entire network is required for a downstream task. Motivated by the recent work on pruning and distilling pre-trained models, we explore strategies to drop layers in pre-trained models, and observe the effect of pruning on downstream GLUE tasks. We were able to prune BERT, RoBERTa and XLNet models up to 40%, while maintaining up to 98% of their original performance. Additionally we show that our pruned models are on par with those built using knowledge distillation, both in terms of size and performance. Our experiments yield interesting observations such as, (i) the lower layers are most critical to maintain downstream task performance, (ii) some tasks such as paraphrase detection and sentence similarity are more robust to the dropping of layers, and (iii) models trained using a different objective function exhibit different learning patterns and w.r.t the layer dropping.

Diffusion Transformers (DiTs) have shown exceptional performance in image generation, yet their large parameter counts incur high computational costs, impeding deployment in resource-constrained settings. To address this, we propose Pluggable Pruning with Contiguous Layer Distillation (PPCL), a flexible structured pruning framework specifically designed for DiT architectures. First, we identify redundant layer intervals through a linear probing mechanism combined with the first-order differential trend analysis of similarity metrics. Subsequently, we propose a plug-and-play teacher-student alternating distillation scheme tailored to integrate depth-wise and width-wise pruning within a single training phase. This distillation framework enables flexible knowledge transfer across diverse pruning ratios, eliminating the need for per-configuration retraining. Extensive experiments on multiple Multi-Modal Diffusion Transformer architecture models demonstrate that PPCL achieves a 50\% reduction in parameter count compared to the full model, with less than 3\% degradation in key objective metrics. Notably, our method maintains high-quality image generation capabilities while achieving higher compression ratios, rendering it well-suited for resource-constrained environments. The open-source code, checkpoints for PPCL can be found at the following link: https://github.com/OPPO-Mente-Lab/Qwen-Image-Pruning.

The inference of transformer-based large language models consists of two sequential stages: 1) a prefilling stage to compute the KV cache of prompts and generate the first token, and 2) a decoding stage to generate subsequent tokens. For long prompts, the KV cache must be computed for all tokens during the prefilling stage, which can significantly increase the time needed to generate the first token. Consequently, the prefilling stage may become a bottleneck in the generation process. An open question remains whether all prompt tokens are essential for generating the first token. To answer this, we introduce a novel method, LazyLLM, that selectively computes the KV for tokens important for the next token prediction in both the prefilling and decoding stages. Contrary to static pruning approaches that prune the prompt at once, LazyLLM allows language models to dynamically select different subsets of tokens from the context in different generation steps, even though they might be pruned in previous steps. Extensive experiments on standard datasets across various tasks demonstrate that LazyLLM is a generic method that can be seamlessly integrated with existing language models to significantly accelerate the generation without fine-tuning. For instance, in the multi-document question-answering task, LazyLLM accelerates the prefilling stage of the LLama 2 7B model by 2.34x while maintaining accuracy.

Transformer-based language models have become a key building block for natural language processing. While these models are extremely accurate, they can be too large and computationally intensive to run on standard deployments. A variety of compression methods, including distillation, quantization, structured and unstructured pruning are known to decrease model size and increase inference speed, with low accuracy loss. In this context, this paper's contributions are two-fold. We perform an in-depth study of the accuracy-compression trade-off for unstructured weight pruning of BERT models. We introduce Optimal BERT Surgeon (oBERT), an efficient and accurate weight pruning method based on approximate second-order information, which we show to yield state-of-the-art results in both stages of language tasks: pre-training and fine-tuning. Specifically, oBERT extends existing work on unstructured second-order pruning by allowing for pruning blocks of weights, and by being applicable at the BERT scale. Second, we investigate the impact of this pruning method when compounding compression approaches to obtain highly compressed but accurate models for deployment on edge devices. These models significantly push boundaries of the current state-of-the-art sparse BERT models with respect to all metrics: model size, inference speed and task accuracy. For example, relative to the dense BERT-base, we obtain 10x model size compression (in MB) with < 1% accuracy drop, 10x CPU-inference speedup with < 2% accuracy drop, and 29x CPU-inference speedup with < 7.5% accuracy drop. Our code, fully integrated with Transformers and SparseML, is available at https://github.com/neuralmagic/sparseml/tree/main/research/optimal_BERT_surgeon_oBERT.

Deployment of Transformer models on edge devices is becoming increasingly challenging due to the exponentially growing inference cost that scales quadratically with the number of tokens in the input sequence. Token pruning is an emerging solution to address this challenge due to its ease of deployment on various Transformer backbones. However, most token pruning methods require computationally expensive fine-tuning, which is undesirable in many edge deployment cases. In this work, we propose Zero-TPrune, the first zero-shot method that considers both the importance and similarity of tokens in performing token pruning. It leverages the attention graph of pre-trained Transformer models to produce an importance distribution for tokens via our proposed Weighted Page Rank (WPR) algorithm. This distribution further guides token partitioning for efficient similarity-based pruning. Due to the elimination of the fine-tuning overhead, Zero-TPrune can prune large models at negligible computational cost, switch between different pruning configurations at no computational cost, and perform hyperparameter tuning efficiently. We evaluate the performance of Zero-TPrune on vision tasks by applying it to various vision Transformer backbones and testing them on ImageNet. Without any fine-tuning, Zero-TPrune reduces the FLOPs cost of DeiT-S by 34.7\% and improves its throughput by 45.3\% with only 0.4\% accuracy loss. Compared with state-of-the-art pruning methods that require fine-tuning, Zero-TPrune not only eliminates the need for fine-tuning after pruning but also does so with only 0.1\% accuracy loss. Compared with state-of-the-art fine-tuning-free pruning methods, Zero-TPrune reduces accuracy loss by up to 49\% with the same or higher throughput.

Self-attention mechanism is the key of the Transformer but often criticized for its computation demands. Previous token pruning works motivate their methods from the view of computation redundancy but still need to load the full network and require same memory costs. This paper introduces a novel strategy that simplifies vision transformers and reduces computational load through the selective removal of non-essential attention layers, guided by entropy considerations. We identify that regarding the attention layer in bottom blocks, their subsequent MLP layers, i.e. two feed-forward layers, can elicit the same entropy quantity. Meanwhile, the accompanied MLPs are under-exploited since they exhibit smaller feature entropy compared to those MLPs in the top blocks. Therefore, we propose to integrate the uninformative attention layers into their subsequent counterparts by degenerating them into identical mapping, yielding only MLP in certain transformer blocks. Experimental results on ImageNet-1k show that the proposed method can remove 40% attention layer of DeiT-B, improving throughput and memory bound without performance compromise. Code is available at https://github.com/sihaoevery/lambda_vit.

Transformers have been successfully applied in the field of video-based 3D human pose estimation. However, the high computational costs of these video pose transformers (VPTs) make them impractical on resource-constrained devices. In this paper, we present a plug-and-play pruning-and-recovering framework, called Hourglass Tokenizer (HoT), for efficient transformer-based 3D human pose estimation from videos. Our HoT begins with pruning pose tokens of redundant frames and ends with recovering full-length tokens, resulting in a few pose tokens in the intermediate transformer blocks and thus improving the model efficiency. To effectively achieve this, we propose a token pruning cluster (TPC) that dynamically selects a few representative tokens with high semantic diversity while eliminating the redundancy of video frames. In addition, we develop a token recovering attention (TRA) to restore the detailed spatio-temporal information based on the selected tokens, thereby expanding the network output to the original full-length temporal resolution for fast inference. Extensive experiments on two benchmark datasets (i.e., Human3.6M and MPI-INF-3DHP) demonstrate that our method can achieve both high efficiency and estimation accuracy compared to the original VPT models. For instance, applying to MotionBERT and MixSTE on Human3.6M, our HoT can save nearly 50% FLOPs without sacrificing accuracy and nearly 40% FLOPs with only 0.2% accuracy drop, respectively. Code and models are available at https://github.com/NationalGAILab/HoT.

Despite achieving remarkable performance on various vision-language tasks, Transformer-based Vision-Language Models (VLMs) suffer from redundancy in inputs and parameters, significantly hampering their efficiency in real-world applications. Moreover, the degree of redundancy in token representations and model parameters, such as attention heads, varies significantly for different inputs. In light of the challenges, we propose SmartTrim, an adaptive acceleration framework for VLMs, which adjusts the computational overhead per instance. Specifically, we integrate lightweight modules into the original backbone to identify and prune redundant token representations and attention heads within each layer. Furthermore, we devise a self-distillation strategy to enhance the consistency between the predictions of the pruned model and its fully-capacity counterpart. Experimental results across various vision-language tasks consistently demonstrate that SmartTrim accelerates the original model by 2-3 times with minimal performance degradation, highlighting the effectiveness and efficiency compared to previous approaches. Code will be available at https://github.com/kugwzk/SmartTrim.

Recent work proposed state-space models (SSMs) as an efficient alternative to transformer-based LLMs. Can these models be pruned to further reduce their computation costs? We adapt several pruning methods to the SSM structure, and apply them to four SSM-based LLMs across multiple tasks. We find that such models are quite robust to some pruning methods (e.g. WANDA), while using other methods lead to fast performance degradation.

Large language models (LLMs) based on transformer are witnessing a notable trend of size expansion, which brings considerable costs to both model training and inference. However, existing methods such as model quantization, knowledge distillation, and model pruning are constrained by various issues, including hardware support limitations, the need for extensive training, and alterations to the internal structure of the model. In this paper, we propose a concise layer-wise pruning method called Layer Collapse (LaCo), in which rear model layers collapse into a prior layer, enabling a rapid reduction in model size while preserving the model structure. Comprehensive experiments show that our method maintains an average task performance of over 80\% at pruning ratios of 25-30\%, significantly outperforming existing state-of-the-art structured pruning methods. We also conduct post-training experiments to confirm that the proposed pruning method effectively inherits the parameters of the original model. Finally, we discuss our motivation from the perspective of layer-wise similarity and evaluate the performance of the pruned LLMs across various pruning ratios.

Transformer-based models, such as BERT and ViT, have achieved state-of-the-art results across different natural language processing (NLP) and computer vision (CV) tasks. However, these models are extremely memory intensive during their fine-tuning process, making them difficult to deploy on GPUs with limited memory resources. To address this issue, we introduce a new tool called SlimFit that reduces the memory requirements of these models by dynamically analyzing their training dynamics and freezing less-contributory layers during fine-tuning. The layers to freeze are chosen using a runtime inter-layer scheduling algorithm. SlimFit adopts quantization and pruning for particular layers to balance the load of dynamic activations and to minimize the memory footprint of static activations, where static activations refer to those that cannot be discarded regardless of freezing. This allows SlimFit to freeze up to 95% of layers and reduce the overall on-device GPU memory usage of transformer-based models such as ViT and BERT by an average of 2.2x, across different NLP and CV benchmarks/datasets such as GLUE, SQuAD 2.0, CIFAR-10, CIFAR-100 and ImageNet with an average degradation of 0.2% in accuracy. For such NLP and CV tasks, SlimFit can reduce up to 3.1x the total on-device memory usage with an accuracy degradation of only up to 0.4%. As a result, while fine-tuning of ViT on ImageNet and BERT on SQuAD 2.0 with a batch size of 128 requires 3 and 2 32GB GPUs respectively, SlimFit enables their fine-tuning on a single 32GB GPU without any significant accuracy degradation.

Large models based on the Transformer architecture play increasingly vital roles in artificial intelligence, particularly within the realms of natural language processing (NLP) and computer vision (CV). Model compression methods reduce their memory and computational cost, which is a necessary step to implement the transformer models on practical devices. Given the unique architecture of transformer, featuring alternative attention and Feedforward Neural Network (FFN) modules, specific compression techniques are required. The efficiency of these compression methods is also paramount, as it is usually impractical to retrain large models on the entire training dataset.This survey provides a comprehensive review of recent compression methods, with a specific focus on their application to transformer models. The compression methods are primarily categorized into pruning, quantization, knowledge distillation, and efficient architecture design. In each category, we discuss compression methods for both CV and NLP tasks, highlighting common underlying principles. At last, we delve into the relation between various compression methods, and discuss the further directions in this domain.

Diffusion Transformer (DiT) has demonstrated remarkable performance in text-to-image generation; however, its large parameter size results in substantial inference overhead. Existing parameter compression methods primarily focus on pruning, but aggressive pruning often leads to severe performance degradation due to reduced model capacity. To address this limitation, we pioneer the transformation of a dense DiT into a Mixture of Experts (MoE) for structured sparsification, reducing the number of activated parameters while preserving model capacity. Specifically, we replace the Feed-Forward Networks (FFNs) in DiT Blocks with MoE layers, reducing the number of activated parameters in the FFNs by 62.5\%. Furthermore, we propose the Mixture of Blocks (MoB) to selectively activate DiT blocks, thereby further enhancing sparsity. To ensure an effective dense-to-MoE conversion, we design a multi-step distillation pipeline, incorporating Taylor metric-based expert initialization, knowledge distillation with load balancing, and group feature loss for MoB optimization. We transform large diffusion transformers (e.g., FLUX.1 [dev]) into an MoE structure, reducing activated parameters by 60\% while maintaining original performance and surpassing pruning-based approaches in extensive experiments. Overall, Dense2MoE establishes a new paradigm for efficient text-to-image generation.

As large language models continue to scale, their growing computational and storage demands pose significant challenges for real-world deployment. In this work, we investigate redundancy within Transformer-based models and propose an entropy-based pruning strategy to enhance efficiency while maintaining performance. Empirical analysis reveals that the entropy of hidden representations decreases in the early blocks but progressively increases across most subsequent blocks. This trend suggests that entropy serves as a more effective measure of information richness within computation blocks. Unlike cosine similarity, which primarily captures geometric relationships, entropy directly quantifies uncertainty and information content, making it a more reliable criterion for pruning. Extensive experiments demonstrate that our entropy-based pruning approach surpasses cosine similarity-based methods in reducing model size while preserving accuracy, offering a promising direction for efficient model deployment.

To remove redundant components of large language models (LLMs) without incurring significant computational costs, this work focuses on single-shot pruning without a retraining phase. We simplify the pruning process for Transformer-based LLMs by identifying a depth-2 pruning structure that functions independently. Additionally, we propose two inference-aware pruning criteria derived from the optimization perspective of output approximation, which outperforms traditional training-aware metrics such as gradient and Hessian. We also introduce a two-step reconstruction technique to mitigate pruning errors without model retraining. Experimental results demonstrate that our approach significantly reduces computational costs and hardware requirements while maintaining superior performance across various datasets and models.

We introduce the One-shot Pruning Technique for Interchangeable Networks (OPTIN) framework as a tool to increase the efficiency of pre-trained transformer architectures without requiring re-training. Recent works have explored improving transformer efficiency, however often incur computationally expensive re-training procedures or depend on architecture-specific characteristics, thus impeding practical wide-scale adoption. To address these shortcomings, the OPTIN framework leverages intermediate feature distillation, capturing the long-range dependencies of model parameters (coined trajectory), to produce state-of-the-art results on natural language, image classification, transfer learning, and semantic segmentation tasks without re-training. Given a FLOP constraint, the OPTIN framework will compress the network while maintaining competitive accuracy performance and improved throughput. Particularly, we show a leq 2% accuracy degradation from NLP baselines and a 0.5% improvement from state-of-the-art methods on image classification at competitive FLOPs reductions. We further demonstrate the generalization of tasks and architecture with comparative performance using Mask2Former for semantic segmentation and cnn-style networks. OPTIN presents one of the first one-shot efficient frameworks for compressing transformer architectures that generalizes well across different class domains, in particular: natural language and image-related tasks, without re-training.

Vision transformers (ViTs) have gained popularity recently. Even without customized image operators such as convolutions, ViTs can yield competitive performance when properly trained on massive data. However, the computational overhead of ViTs remains prohibitive, due to stacking multi-head self-attention modules and else. Compared to the vast literature and prevailing success in compressing convolutional neural networks, the study of Vision Transformer compression has also just emerged, and existing works focused on one or two aspects of compression. This paper proposes a unified ViT compression framework that seamlessly assembles three effective techniques: pruning, layer skipping, and knowledge distillation. We formulate a budget-constrained, end-to-end optimization framework, targeting jointly learning model weights, layer-wise pruning ratios/masks, and skip configurations, under a distillation loss. The optimization problem is then solved using the primal-dual algorithm. Experiments are conducted with several ViT variants, e.g. DeiT and T2T-ViT backbones on the ImageNet dataset, and our approach consistently outperforms recent competitors. For example, DeiT-Tiny can be trimmed down to 50\% of the original FLOPs almost without losing accuracy. Codes are available online:~https://github.com/VITA-Group/UVC.

We introduce ReplaceMe, a generalized training-free depth pruning method that effectively replaces transformer blocks with a linear operation, while maintaining high performance for low compression ratios. In contrast to conventional pruning approaches that require additional training or fine-tuning, our approach requires only a small calibration dataset that is used to estimate a linear transformation to approximate the pruned blocks. This estimated linear mapping can be seamlessly merged with the remaining transformer blocks, eliminating the need for any additional network parameters. Our experiments show that ReplaceMe consistently outperforms other training-free approaches and remains highly competitive with state-of-the-art pruning methods that involve extensive retraining/fine-tuning and architectural modifications. Applied to several large language models (LLMs), ReplaceMe achieves up to 25% pruning while retaining approximately 90% of the original model's performance on open benchmarks - without any training or healing steps, resulting in minimal computational overhead (see Fig.1). We provide an open-source library implementing ReplaceMe alongside several state-of-the-art depth pruning techniques, available at this repository.

Large language models (LLMs) have revolutionized AI applications, yet their high computational and memory demands hinder their widespread deployment. Existing compression techniques focus on intra-block optimizations (e.g. low-rank approximation, attention head pruning), while the repetitive layered structure of transformers implies significant inter-block redundancy - a dimension largely unexplored beyond key-value (KV) caching. Inspired by dictionary learning in CNNs, we propose a framework for structured weight sharing across transformer layers. Our approach decomposes attention projection matrices into shared dictionary atoms, reducing the attention module's parameters by 66.7% while achieving on-par performance. Unlike complex methods requiring distillation or architectural changes, MASA (Matrix Atom Sharing in Attention) operates as a drop-in replacement - trained with standard optimizers - and represents each layer's weights as linear combinations of shared matrix atoms. Experiments across scales (100M-700M parameters) show that MASA achieves better benchmark accuracy and perplexity than grouped-query attention (GQA), low-rank baselines and recently proposed Repeat-all-over/Sequential sharing at comparable parameter budgets. Ablation studies confirm robustness to the dictionary size and the efficacy of shared representations in capturing cross-layer statistical regularities. Extending to Vision Transformers (ViT), MASA matches performance metrics on image classification and detection tasks with 66.7% fewer attention parameters. By combining dictionary learning strategies with transformer efficiency, MASA offers a scalable blueprint for parameter-efficient models without sacrificing performance. Finally, we investigate the possibility of employing MASA on pretrained LLMs to reduce their number of parameters without experiencing any significant drop in their performance.

Autoregressive Language Models (LLMs) trained on Next-Token Prediction (NTP) often suffer from ``Topic Drift'' where the generation wanders away from the initial prompt due to a reliance on local associations rather than global planning holtzman2019curious. While scaling model size mitigates this brown2020language, the fundamental myopia of the NTP objective remains. In this work, we introduce the Idea-Gated Transformer, a novel architecture that separates semantic planning from syntactic generation. We introduce an auxiliary ``Idea Head'' trained to predict the bag-of-words distribution for a future context window, creating a latent ``Concept Vector'' that actively gates the main vocabulary during generation. We propose a differentiable gating mechanism that suppresses semantically irrelevant tokens, effectively pruning the search space in real-time. Experiments on WikiText-103 demonstrate that while the Idea-Gated model achieves comparable validation perplexity to a standard GPT-2 baseline, it exhibits significantly superior Domain Retention. Qualitative and quantitative analysis reveals that the gating mechanism successfully locks generation into specific semantic clusters (e.g., Finance, Science) and resists associative drift, offering a parameter-efficient path toward more controllable language modeling.

Pruning large language models (LLMs) is a promising solution for reducing model sizes and computational complexity while preserving performance. Traditional layer-wise pruning methods often adopt a uniform sparsity approach across all layers, which leads to suboptimal performance due to the varying significance of individual transformer layers within the model not being accounted for. To this end, we propose the Shapley Value-based Non-Uniform Pruning (SV-NUP) method for LLMs. This approach quantifies the contribution of each transformer layer to the overall model performance, enabling the assignment of tailored pruning budgets to different layers to retain critical parameters. To further improve efficiency, we design the Sliding Window-based Shapley Value approximation method. It substantially reduces computational overhead compared to exact SV calculation methods. Extensive experiments on various LLMs including LLaMA-v1, LLaMA-v2 and OPT demonstrate the effectiveness of the proposed approach. The results reveal that non-uniform pruning significantly enhances the performance of pruned models. Notably, SV-NUP achieves a reduction in perplexity (PPL) of 18.01% and 19.55% on LLaMA-7B and LLaMA-13B, respectively, compared to SparseGPT at 70% sparsity.

Vision transformers have demonstrated remarkable success in a wide range of computer vision tasks over the last years. However, their high computational costs remain a significant barrier to their practical deployment. In particular, the complexity of transformer models is quadratic with respect to the number of input tokens. Therefore techniques that reduce the number of input tokens that need to be processed have been proposed. This paper introduces Learned Thresholds token Merging and Pruning (LTMP), a novel approach that leverages the strengths of both token merging and token pruning. LTMP uses learned threshold masking modules that dynamically determine which tokens to merge and which to prune. We demonstrate our approach with extensive experiments on vision transformers on the ImageNet classification task. Our results demonstrate that LTMP achieves state-of-the-art accuracy across reduction rates while requiring only a single fine-tuning epoch, which is an order of magnitude faster than previous methods. Code is available at https://github.com/Mxbonn/ltmp .

Long sequences of text are challenging in the context of transformers, due to quadratic memory increase in the self-attention mechanism. As this issue directly affects the translation from natural language to SQL queries (as techniques usually take as input a concatenated text with the question and the database schema), we present techniques that allow long text sequences to be handled by transformers with up to 512 input tokens. We propose a training process with database schema pruning (removal of tables and columns names that are useless for the query of interest). In addition, we used a multilingual approach with the mT5-large model fine-tuned with a data-augmented Spider dataset in four languages simultaneously: English, Portuguese, Spanish, and French. Our proposed technique used the Spider dataset and increased the exact set match accuracy results from 0.718 to 0.736 in a validation dataset (Dev). Source code, evaluations, and checkpoints are available at: https://github.com/C4AI/gap-text2sql.

The transformer extends its success from the language to the vision domain. Because of the stacked self-attention and cross-attention blocks, the acceleration deployment of vision transformer on GPU hardware is challenging and also rarely studied. This paper thoroughly designs a compression scheme to maximally utilize the GPU-friendly 2:4 fine-grained structured sparsity and quantization. Specially, an original large model with dense weight parameters is first pruned into a sparse one by 2:4 structured pruning, which considers the GPU's acceleration of 2:4 structured sparse pattern with FP16 data type, then the floating-point sparse model is further quantized into a fixed-point one by sparse-distillation-aware quantization aware training, which considers GPU can provide an extra speedup of 2:4 sparse calculation with integer tensors. A mixed-strategy knowledge distillation is used during the pruning and quantization process. The proposed compression scheme is flexible to support supervised and unsupervised learning styles. Experiment results show GPUSQ-ViT scheme achieves state-of-the-art compression by reducing vision transformer models 6.4-12.7 times on model size and 30.3-62 times on FLOPs with negligible accuracy degradation on ImageNet classification, COCO detection and ADE20K segmentation benchmarking tasks. Moreover, GPUSQ-ViT can boost actual deployment performance by 1.39-1.79 times and 3.22-3.43 times of latency and throughput on A100 GPU, and 1.57-1.69 times and 2.11-2.51 times improvement of latency and throughput on AGX Orin.

Structured pruning and quantization are fundamental techniques used to reduce the size of deep neural networks (DNNs) and typically are applied independently. Applying these techniques jointly via co-optimization has the potential to produce smaller, high-quality models. However, existing joint schemes are not widely used because of (1) engineering difficulties (complicated multi-stage processes), (2) black-box optimization (extensive hyperparameter tuning to control the overall compression), and (3) insufficient architecture generalization. To address these limitations, we present the framework GETA, which automatically and efficiently performs joint structured pruning and quantization-aware training on any DNNs. GETA introduces three key innovations: (i) a quantization-aware dependency graph (QADG) that constructs a pruning search space for generic quantization-aware DNN, (ii) a partially projected stochastic gradient method that guarantees layerwise bit constraints are satisfied, and (iii) a new joint learning strategy that incorporates interpretable relationships between pruning and quantization. We present numerical experiments on both convolutional neural networks and transformer architectures that show that our approach achieves competitive (often superior) performance compared to existing joint pruning and quantization methods.

Large language models (LLMs) with billions of parameters demonstrate impressive performance. However, the widely used Multi-Head Attention (MHA) in LLMs incurs substantial computational and memory costs during inference. While some efforts have optimized attention mechanisms by pruning heads or sharing parameters among heads, these methods often lead to performance degradation or necessitate substantial continued pre-training costs to restore performance. Based on the analysis of attention redundancy, we design a Decoupled-Head Attention (DHA) mechanism. DHA adaptively configures group sharing for key heads and value heads across various layers, achieving a better balance between performance and efficiency. Inspired by the observation of clustering similar heads, we propose to progressively transform the MHA checkpoint into the DHA model through linear fusion of similar head parameters step by step, retaining the parametric knowledge of the MHA checkpoint. We construct DHA models by transforming various scales of MHA checkpoints given target head budgets. Our experiments show that DHA remarkably requires a mere 0.25\% of the original model's pre-training budgets to achieve 97.6\% of performance while saving 75\% of KV cache. Compared to Group-Query Attention (GQA), DHA achieves a 5times training acceleration, a maximum of 13.93\% performance improvement under 0.01\% pre-training budget, and 4\% relative improvement under 0.05\% pre-training budget.

Vision Transformers (ViTs) are normally regarded as a stack of transformer layers. In this work, we propose a novel view of ViTs showing that they can be seen as ensemble networks containing multiple parallel paths with different lengths. Specifically, we equivalently transform the traditional cascade of multi-head self-attention (MSA) and feed-forward network (FFN) into three parallel paths in each transformer layer. Then, we utilize the identity connection in our new transformer form and further transform the ViT into an explicit multi-path ensemble network. From the new perspective, these paths perform two functions: the first is to provide the feature for the classifier directly, and the second is to provide the lower-level feature representation for subsequent longer paths. We investigate the influence of each path for the final prediction and discover that some paths even pull down the performance. Therefore, we propose the path pruning and EnsembleScale skills for improvement, which cut out the underperforming paths and re-weight the ensemble components, respectively, to optimize the path combination and make the short paths focus on providing high-quality representation for subsequent paths. We also demonstrate that our path combination strategies can help ViTs go deeper and act as high-pass filters to filter out partial low-frequency signals. To further enhance the representation of paths served for subsequent paths, self-distillation is applied to transfer knowledge from the long paths to the short paths. This work calls for more future research to explain and design ViTs from new perspectives.

This paper presents SANA-1.5, a linear Diffusion Transformer for efficient scaling in text-to-image generation. Building upon SANA-1.0, we introduce three key innovations: (1) Efficient Training Scaling: A depth-growth paradigm that enables scaling from 1.6B to 4.8B parameters with significantly reduced computational resources, combined with a memory-efficient 8-bit optimizer. (2) Model Depth Pruning: A block importance analysis technique for efficient model compression to arbitrary sizes with minimal quality loss. (3) Inference-time Scaling: A repeated sampling strategy that trades computation for model capacity, enabling smaller models to match larger model quality at inference time. Through these strategies, SANA-1.5 achieves a text-image alignment score of 0.72 on GenEval, which can be further improved to 0.80 through inference scaling, establishing a new SoTA on GenEval benchmark. These innovations enable efficient model scaling across different compute budgets while maintaining high quality, making high-quality image generation more accessible.

Large language models (LLMs) have demonstrated outstanding performance in various tasks, such as text summarization, text question-answering, and etc. While their performance is impressive, the computational footprint due to their vast number of parameters can be prohibitive. Existing solutions such as SparseGPT and Wanda attempt to alleviate this issue through weight pruning. However, their layer-wise approach results in significant perturbation to the model's output and requires meticulous hyperparameter tuning, such as the pruning rate, which can adversely affect overall model performance. To address this, this paper introduces a novel LLM pruning technique dubbed blockwise parameter-efficient sparsity allocation (BESA) by applying a blockwise reconstruction loss. In contrast to the typical layer-wise pruning techniques, BESA is characterized by two distinctive attributes: i) it targets the overall pruning error with respect to individual transformer blocks, and ii) it allocates layer-specific sparsity in a differentiable manner, both of which ensure reduced performance degradation after pruning. Our experiments show that BESA achieves state-of-the-art performance, efficiently pruning LLMs like LLaMA1, and LLaMA2 with 7B to 70B parameters on a single A100 GPU in just five hours. Code is available at https://github.com/OpenGVLab/LLMPrune-BESA{here}.

With the rapid growth in the size and complexity of large language models (LLMs), the costs associated with their training and inference have escalated significantly. Research indicates that certain layers in LLMs harbor substantial redundancy, and pruning these layers has minimal impact on the overall performance. While various layer pruning methods have been developed based on this insight, they generally overlook the finer-grained redundancies within the layers themselves. In this paper, we delve deeper into the architecture of LLMs and demonstrate that finer-grained pruning can be achieved by targeting redundancies in multi-head attention (MHA) and multi-layer perceptron (MLP) blocks. We propose a novel, training-free structured pruning approach called BlockPruner. Unlike existing layer pruning methods, BlockPruner segments each Transformer layer into MHA and MLP blocks. It then assesses the importance of these blocks using perplexity measures and applies a heuristic search for iterative pruning. We applied BlockPruner to LLMs of various sizes and architectures and validated its performance across a wide range of downstream tasks. Experimental results show that BlockPruner achieves more granular and effective pruning compared to state-of-the-art baselines.

Transformer-based end-to-end speech recognition has achieved great success. However, the large footprint and computational overhead make it difficult to deploy these models in some real-world applications. Model compression techniques can reduce the model size and speed up inference, but the compressed model has a fixed architecture which might be suboptimal. We propose a novel Transformer encoder with Input-Dependent Dynamic Depth (I3D) to achieve strong performance-efficiency trade-offs. With a similar number of layers at inference time, I3D-based models outperform the vanilla Transformer and the static pruned model via iterative layer pruning. We also present interesting analysis on the gate probabilities and the input-dependency, which helps us better understand deep encoders.

Adjusting the latency, power, and accuracy of natural language understanding models is a desirable objective of efficient architecture development. This paper proposes an efficient transformer architecture that adjusts the inference computational cost adaptively with desired inference latency speedup. The proposed encoder model can work with fewer Floating Point Operations (FLOPs) than the original Transformer architecture. In fine-tuning phase, the proposed method detects more important hidden sequence elements (word-vectors) in each encoder layer by a proposed Attention Context Contribution (ACC) metric. It eliminates the less important word-vectors based on a new strategy. A mathematical inference speedup analysis is proposed to estimate the speedup accurately to adjust the latency and computational cost of fine-tuning and inference phases. After the fine-tuning phase, by the method offline-tuning property, the inference latency of the model can be adjusted in a wide range of inference speedup selections. The proposed method is applied to the BERTbase model for evaluation. Extensive experiments show that most of the word-vectors in higher BERT encoder layers have less contribution to the subsequent layers; hence, they can be eliminated to improve the inference latency. Experimental results on extensive sentiment analysis, classification, and regression benchmarks like GLUE showed that the method is effective in various datasets. The proposed method improves the inference latency of BERTbase by up to 4.8 times with less than 0.75% accuracy drop on average.

Transformer models have achieved promising results on natural language processing (NLP) tasks including extractive question answering (QA). Common Transformer encoders used in NLP tasks process the hidden states of all input tokens in the context paragraph throughout all layers. However, different from other tasks such as sequence classification, answering the raised question does not necessarily need all the tokens in the context paragraph. Following this motivation, we propose Block-skim, which learns to skim unnecessary context in higher hidden layers to improve and accelerate the Transformer performance. The key idea of Block-Skim is to identify the context that must be further processed and those that could be safely discarded early on during inference. Critically, we find that such information could be sufficiently derived from the self-attention weights inside the Transformer model. We further prune the hidden states corresponding to the unnecessary positions early in lower layers, achieving significant inference-time speedup. To our surprise, we observe that models pruned in this way outperform their full-size counterparts. Block-Skim improves QA models' accuracy on different datasets and achieves 3 times speedup on BERT-base model.

Large-scale fine-grained image retrieval (FGIR) aims to retrieve images belonging to the same subcategory as a given query by capturing subtle differences in a large-scale setting. Recently, Vision Transformers (ViT) have been employed in FGIR due to their powerful self-attention mechanism for modeling long-range dependencies. However, most Transformer-based methods focus primarily on leveraging self-attention to distinguish fine-grained details, while overlooking the high computational complexity and redundant dependencies inherent to these models, limiting their scalability and effectiveness in large-scale FGIR. In this paper, we propose an Efficient and Effective ViT-based framework, termed EET, which integrates token pruning module with a discriminative transfer strategy to address these limitations. Specifically, we introduce a content-based token pruning scheme to enhance the efficiency of the vanilla ViT, progressively removing background or low-discriminative tokens at different stages by exploiting feature responses and self-attention mechanism. To ensure the resulting efficient ViT retains strong discriminative power, we further present a discriminative transfer strategy comprising both discriminative knowledge transfer and discriminative region guidance. Using a distillation paradigm, these components transfer knowledge from a larger ``teacher'' ViT to a more efficient ``student'' model, guiding the latter to focus on subtle yet crucial regions in a cost-free manner. Extensive experiments on two widely-used fine-grained datasets and four large-scale fine-grained datasets demonstrate the effectiveness of our method. Specifically, EET reduces the inference latency of ViT-Small by 42.7\% and boosts the retrieval performance of 16-bit hash codes by 5.15\% on the challenging NABirds dataset.

Diffusion Transformers (DiT) deliver impressive generative performance but face prohibitive computational demands due to both the quadratic complexity of token-based self-attention and the need for extensive sampling steps. While recent research has focused on accelerating sampling, the structural inefficiencies of DiT remain underexplored. We propose FlexDiT, a framework that dynamically adapts token density across both spatial and temporal dimensions to achieve computational efficiency without compromising generation quality. Spatially, FlexDiT employs a three-segment architecture that allocates token density based on feature requirements at each layer: Poolingformer in the bottom layers for efficient global feature extraction, Sparse-Dense Token Modules (SDTM) in the middle layers to balance global context with local detail, and dense tokens in the top layers to refine high-frequency details. Temporally, FlexDiT dynamically modulates token density across denoising stages, progressively increasing token count as finer details emerge in later timesteps. This synergy between FlexDiT's spatially adaptive architecture and its temporal pruning strategy enables a unified framework that balances efficiency and fidelity throughout the generation process. Our experiments demonstrate FlexDiT's effectiveness, achieving a 55% reduction in FLOPs and a 175% improvement in inference speed on DiT-XL with only a 0.09 increase in FID score on 512times512 ImageNet images, a 56% reduction in FLOPs across video generation datasets including FaceForensics, SkyTimelapse, UCF101, and Taichi-HD, and a 69% improvement in inference speed on PixArt-alpha on text-to-image generation task with a 0.24 FID score decrease. FlexDiT provides a scalable solution for high-quality diffusion-based generation compatible with further sampling optimization techniques.

Structured pruning fundamentally reduces computational and memory overheads of large language models (LLMs) and offers a feasible solution for end-side LLM deployment. Structurally pruned models remain dense and high-precision, highly compatible with further tuning and compression. However, as the coarse-grained structured pruning poses large damage to the highly interconnected model, achieving a high compression ratio for scaled-up LLMs remains a challenge. In this paper, we introduce a task-agnostic structured pruning approach coupled with a compact Transformer architecture design. The proposed approach, named TransAct, reduces transitional activations inside multi-head attention (MHA) and multi-layer perceptron (MLP) modules, while preserving the inter-module activations that are sensitive to perturbations. Hence, the LLM is pruned into an intra-module low-rank architecture, significantly reducing weights, KV Cache and attention computation. TransAct is implemented on the LLaMA model and evaluated on downstream benchmarks. Results verify the optimality of our approach at high compression with respect to both efficiency and performance. Further, ablation studies reveal the strength of activation-guided iterative pruning and provide experimental analysis on the redundancy of MHA and MLP modules.

High-resolution images enable neural networks to learn richer visual representations. However, this improved performance comes at the cost of growing computational complexity, hindering their usage in latency-sensitive applications. As not all pixels are equal, skipping computations for less-important regions offers a simple and effective measure to reduce the computation. This, however, is hard to be translated into actual speedup for CNNs since it breaks the regularity of the dense convolution workload. In this paper, we introduce SparseViT that revisits activation sparsity for recent window-based vision transformers (ViTs). As window attentions are naturally batched over blocks, actual speedup with window activation pruning becomes possible: i.e., ~50% latency reduction with 60% sparsity. Different layers should be assigned with different pruning ratios due to their diverse sensitivities and computational costs. We introduce sparsity-aware adaptation and apply the evolutionary search to efficiently find the optimal layerwise sparsity configuration within the vast search space. SparseViT achieves speedups of 1.5x, 1.4x, and 1.3x compared to its dense counterpart in monocular 3D object detection, 2D instance segmentation, and 2D semantic segmentation, respectively, with negligible to no loss of accuracy.

Growing context lengths in transformer-based language models have made the key-value (KV) cache a critical inference bottleneck. While many KV cache pruning methods have been proposed, they have not yet been adopted in major inference engines due to speed--accuracy trade-offs. We introduce KVzap, a fast, input-adaptive approximation of KVzip that works in both prefilling and decoding. On Qwen3-8B, Llama-3.1-8B-Instruct, and Qwen3-32B across long-context and reasoning tasks, KVzap achieves 2--4times KV cache compression with negligible accuracy loss and achieves state-of-the-art performance on the KVpress leaderboard. Code and models are available at https://github.com/NVIDIA/kvpress.

In recent years, the long-range attention mechanism of vision transformers has driven significant performance breakthroughs across various computer vision tasks. However, the traditional self-attention mechanism, which processes both informative and non-informative tokens, suffers from inefficiency and inaccuracies. While sparse attention mechanisms have been introduced to mitigate these issues by pruning tokens involved in attention, they often lack context-awareness and intelligence. These mechanisms frequently apply a uniform token selection strategy across different inputs for batch training or optimize efficiency only for the inference stage. To overcome these challenges, we propose a novel algorithm: Select and Pack Attention (SPA). SPA dynamically selects informative tokens using a low-cost gating layer supervised by selection labels and packs these tokens into new batches, enabling a variable number of tokens to be used in parallelized GPU batch training and inference. Extensive experiments across diverse datasets and computer vision tasks demonstrate that SPA delivers superior performance and efficiency, including a 0.6 mAP improvement in object detection and a 16.4% reduction in computational costs.

The increasing size of language models raises great research interests in parameter-efficient fine-tuning such as LoRA that freezes the pre-trained model, and injects small-scale trainable parameters for multiple downstream tasks (e.g., summarization, question answering and translation). To further enhance the efficiency of fine-tuning, we propose a framework that integrates LoRA and structured layer pruning. The integrated framework is validated on two created deidentified medical report summarization datasets based on MIMIC-IV-Note and two public medical dialogue datasets. By tuning 0.6% parameters of the original model and pruning over 30% Transformer-layers, our framework can reduce 50% of GPU memory usage and speed up 100% of the training phase, while preserving over 92% generation qualities on free-text sequence-to-sequence tasks.

Learning-based 3D visual geometry models have benefited substantially from large-scale transformers. Among these, StreamVGGT leverages frame-wise causal attention for strong streaming reconstruction, but suffers from unbounded KV cache growth, leading to escalating memory consumption and inference latency as input frames accumulate. We propose XStreamVGGT, a tuning-free approach that systematically compresses the KV cache through joint pruning and quantization, enabling extremely memory-efficient streaming inference. Specifically, redundant KVs originating from multi-view inputs are pruned through efficient token importance identification, enabling a fixed memory budget. Leveraging the unique distribution of KV tensors, we incorporate KV quantization to further reduce memory consumption. Extensive evaluations show that XStreamVGGT achieves mostly negligible performance degradation while substantially reducing memory usage by 4.42times and accelerating inference by 5.48times, enabling scalable and practical streaming 3D applications. The code is available at https://github.com/ywh187/XStreamVGGT/.

Vision transformers have been widely explored in various vision tasks. Due to heavy computational cost, much interest has aroused for compressing vision transformer dynamically in the aspect of tokens. Current methods mainly pay attention to token pruning or merging to reduce token numbers, in which tokens are compressed exclusively, causing great information loss and therefore post-training is inevitably required to recover the performance. In this paper, we rethink token reduction and unify the process as an explicit form of token matrix transformation, in which all existing methods are constructing special forms of matrices within the framework. Furthermore, we propose a many-to-many Token Transforming framework that serves as a generalization of all existing methods and reserves the most information, even enabling training-free acceleration. We conduct extensive experiments to validate our framework. Specifically, we reduce 40% FLOPs and accelerate DeiT-S by times1.5 with marginal 0.1% accuracy drop. Furthermore, we extend the method to dense prediction tasks including segmentation, object detection, depth estimation, and language model generation. Results demonstrate that the proposed method consistently achieves substantial improvements, offering a better computation-performance trade-off, impressive budget reduction and inference acceleration.

Deep learning models are increasingly deployed on resource-constrained edge devices for real-time data analytics. In recent years, Vision Transformer models and their variants have demonstrated outstanding performance across various computer vision tasks. However, their high computational demands and inference latency pose significant challenges for model deployment on resource-constraint edge devices. To address this issue, we propose a novel Vision Transformer splitting framework, ED-ViT, designed to execute complex models across multiple edge devices efficiently. Specifically, we partition Vision Transformer models into several sub-models, where each sub-model is tailored to handle a specific subset of data classes. To further minimize computation overhead and inference latency, we introduce a class-wise pruning technique that reduces the size of each sub-model. We conduct extensive experiments on five datasets with three model structures, demonstrating that our approach significantly reduces inference latency on edge devices and achieves a model size reduction of up to 28.9 times and 34.1 times, respectively, while maintaining test accuracy comparable to the original Vision Transformer. Additionally, we compare ED-ViT with two state-of-the-art methods that deploy CNN and SNN models on edge devices, evaluating accuracy, inference time, and overall model size. Our comprehensive evaluation underscores the effectiveness of the proposed ED-ViT framework.

Sparse Mixture of Expert (SMoE) models have emerged as a scalable alternative to dense models in language modeling. These models use conditionally activated feedforward subnetworks in transformer blocks, allowing for a separation between total model parameters and per-example computation. However, large token-routed SMoE models face a significant challenge: during inference, the entire model must be used for a sequence or a batch, resulting in high latencies in a distributed setting that offsets the advantages of per-token sparse activation. Our research explores task-specific model pruning to inform decisions about designing SMoE architectures, mainly modulating the choice of expert counts in pretraining. We investigate whether such pruned models offer advantages over smaller SMoE models trained from scratch, when evaluating and comparing them individually on tasks. To that end, we introduce an adaptive task-aware pruning technique UNCURL to reduce the number of experts per MoE layer in an offline manner post-training. Our findings reveal a threshold pruning factor for the reduction that depends on the number of experts used in pretraining, above which, the reduction starts to degrade model performance. These insights contribute to our understanding of model design choices when pretraining with SMoE architectures, particularly useful when considering task-specific inference optimization for later stages.

Recent years have seen a phenomenal rise in performance and applications of transformer neural networks. The family of transformer networks, including Bidirectional Encoder Representations from Transformer (BERT), Generative Pretrained Transformer (GPT) and Vision Transformer (ViT), have shown their effectiveness across Natural Language Processing (NLP) and Computer Vision (CV) domains. Transformer-based networks such as ChatGPT have impacted the lives of common men. However, the quest for high predictive performance has led to an exponential increase in transformers' memory and compute footprint. Researchers have proposed techniques to optimize transformer inference at all levels of abstraction. This paper presents a comprehensive survey of techniques for optimizing the inference phase of transformer networks. We survey techniques such as knowledge distillation, pruning, quantization, neural architecture search and lightweight network design at the algorithmic level. We further review hardware-level optimization techniques and the design of novel hardware accelerators for transformers. We summarize the quantitative results on the number of parameters/FLOPs and accuracy of several models/techniques to showcase the tradeoff exercised by them. We also outline future directions in this rapidly evolving field of research. We believe that this survey will educate both novice and seasoned researchers and also spark a plethora of research efforts in this field.

Overparameterized transformer networks have obtained state of the art results in various natural language processing tasks, such as machine translation, language modeling, and question answering. These models contain hundreds of millions of parameters, necessitating a large amount of computation and making them prone to overfitting. In this work, we explore LayerDrop, a form of structured dropout, which has a regularization effect during training and allows for efficient pruning at inference time. In particular, we show that it is possible to select sub-networks of any depth from one large network without having to finetune them and with limited impact on performance. We demonstrate the effectiveness of our approach by improving the state of the art on machine translation, language modeling, summarization, question answering, and language understanding benchmarks. Moreover, we show that our approach leads to small BERT-like models of higher quality compared to training from scratch or using distillation.

The grand vision of enabling persistent, large-scale 3D visual geometry understanding is shackled by the irreconcilable demands of scalability and long-term stability. While offline models like VGGT achieve inspiring geometry capability, their batch-based nature renders them irrelevant for live systems. Streaming architectures, though the intended solution for live operation, have proven inadequate. Existing methods either fail to support truly infinite-horizon inputs or suffer from catastrophic drift over long sequences. We shatter this long-standing dilemma with InfiniteVGGT, a causal visual geometry transformer that operationalizes the concept of a rolling memory through a bounded yet adaptive and perpetually expressive KV cache. Capitalizing on this, we devise a training-free, attention-agnostic pruning strategy that intelligently discards obsolete information, effectively ``rolling'' the memory forward with each new frame. Fully compatible with FlashAttention, InfiniteVGGT finally alleviates the compromise, enabling infinite-horizon streaming while outperforming existing streaming methods in long-term stability. The ultimate test for such a system is its performance over a truly infinite horizon, a capability that has been impossible to rigorously validate due to the lack of extremely long-term, continuous benchmarks. To address this critical gap, we introduce the Long3D benchmark, which, for the first time, enables a rigorous evaluation of continuous 3D geometry estimation on sequences about 10,000 frames. This provides the definitive evaluation platform for future research in long-term 3D geometry understanding. Code is available at: https://github.com/AutoLab-SAI-SJTU/InfiniteVGGT

In recent years, Transformer-based language models have become the standard approach for natural language processing tasks. However, stringent throughput and latency requirements in industrial applications are limiting their adoption. To mitigate the gap, model compression techniques such as structured pruning are being used to improve inference efficiency. However, most existing neural network inference runtimes lack adequate support for structured sparsity. In this paper, we propose an efficient sparse deep learning inference software stack for Transformer-based language models where the weights are pruned with constant block size. Our sparse software accelerator leverages Intel Deep Learning Boost to maximize the performance of sparse matrix - dense matrix multiplication (commonly abbreviated as SpMM) on CPUs. Our SpMM kernel outperforms the existing sparse libraries (oneMKL, TVM, and LIBXSMM) by an order of magnitude on a wide range of GEMM shapes under 5 representative sparsity ratios (70%, 75%, 80%, 85%, 90%). Moreover, our SpMM kernel shows up to 5x speedup over dense GEMM kernel of oneDNN, a well-optimized dense library widely used in industry. We apply our sparse accelerator on widely-used Transformer-based language models including Bert-Mini, DistilBERT, Bert-Base, and BERT-Large. Our sparse inference software shows up to 1.5x speedup over Neural Magic's Deepsparse under same configurations on Xeon on Amazon Web Services under proxy production latency constraints. We also compare our solution with two framework-based inference solutions, ONNX Runtime and PyTorch, and demonstrate up to 37x speedup over ONNX Runtime and 345x over PyTorch on Xeon under the latency constraints. All the source code is publicly available on Github: https://github.com/intel/intel-extension-for-transformers.

The extensive memory footprint of pre-trained language models (PLMs) can hinder deployment in memory-constrained settings, such as cloud environments or on-device. PLMs use embedding matrices to represent extensive vocabularies, forming a large proportion of the model parameters. While previous work towards parameter-efficient PLM development has considered pruning parameters within the transformer layers, pruning the embedding matrix as part of fine-tuning or inference has yet to be explored. We first demonstrate that a significant proportion of the vocabulary remains unused in these scenarios. We then propose a simple yet effective approach that leverages this finding to minimize the memory footprint of the embedding matrix. We show that this approach provides substantial reductions in memory usage across a wide range of models and tasks. Notably, our approach maintains equivalent downstream task performance while allowing a more efficient use of compute resources.

We propose a novel Two-Stage framework for Structured Pruning (2SSP) for pruning Large Language Models (LLMs), which combines two different strategies of pruning, namely Width and Depth Pruning. The first stage (Width Pruning) removes entire neurons, hence their corresponding rows and columns, aiming to preserve the connectivity among the pruned structures in the intermediate state of the Feed-Forward Networks in each Transformer block. This is done based on an importance score measuring the impact of each neuron over the output magnitude. The second stage (Depth Pruning), instead, removes entire Attention submodules. This is done by applying an iterative process that removes the Attention submodules with the minimum impact on a given metric of interest (in our case, perplexity). We also propose a novel mechanism to balance the sparsity rate of the two stages w.r.t. to the desired global sparsity. We test 2SSP on four LLM families and three sparsity rates (25\%, 37.5\%, and 50\%), measuring the resulting perplexity over three language modeling datasets as well as the performance over six downstream tasks. Our method consistently outperforms five state-of-the-art competitors over three language modeling and six downstream tasks, with an up to two-order-of-magnitude gain in terms of pruning time. The code is available at available at https://github.com/FabrizioSandri/2SSP.

Real-world data contains a vast amount of multimodal information, among which vision and language are the two most representative modalities. Moreover, increasingly heavier models, e.g., Transformers, have attracted the attention of researchers to model compression. However, how to compress multimodal models, especially vison-language Transformers, is still under-explored. This paper proposes the Unified and Progressive Pruning (\emph{UPop}) as a universal vison-language Transformer compression framework, which incorporates 1) unifiedly searching multimodal subnets in a continuous optimization space from the original model, which enables automatic assignment of pruning ratios among compressible modalities and structures; 2) progressively searching and retraining the subnet, which maintains convergence between the search and retrain to attain higher compression ratios. Experiments on various tasks, datasets, and model architectures demonstrate the effectiveness and versatility of the proposed UPop framework. The code is available at https://github.com/sdc17/UPop.

The increasing size of large language models (LLMs) has introduced challenges in their training and inference. Removing model components is perceived as a solution to tackle the large model sizes, however, existing pruning methods solely focus on performance, without considering an essential aspect for the responsible use of LLMs: model fairness. It is crucial to address the fairness of LLMs towards diverse groups, such as women, Black people, LGBTQ+, Jewish communities, among others, as they are being deployed and available to a wide audience. In this work, first, we investigate how attention heads impact fairness and performance in pre-trained transformer-based language models. We then propose a novel method to prune the attention heads that negatively impact fairness while retaining the heads critical for performance, i.e. language modeling capabilities. Our approach is practical in terms of time and resources, as it does not require fine-tuning the final pruned, and fairer, model. Our findings demonstrate a reduction in gender bias by 19%, 19.5%, 39.5%, 34.7%, 23%, and 8% for DistilGPT-2, GPT-2, GPT-Neo of two different sizes, GPT-J, and Llama 2 models, respectively, in comparison to the biased model, with only a slight decrease in performance.

Structural pruning enables model acceleration by removing structurally-grouped parameters from neural networks. However, the parameter-grouping patterns vary widely across different models, making architecture-specific pruners, which rely on manually-designed grouping schemes, non-generalizable to new architectures. In this work, we study a highly-challenging yet barely-explored task, any structural pruning, to tackle general structural pruning of arbitrary architecture like CNNs, RNNs, GNNs and Transformers. The most prominent obstacle towards this goal lies in the structural coupling, which not only forces different layers to be pruned simultaneously, but also expects all removed parameters to be consistently unimportant, thereby avoiding structural issues and significant performance degradation after pruning. To address this problem, we propose a general and {fully automatic} method, Dependency Graph (DepGraph), to explicitly model the dependency between layers and comprehensively group coupled parameters for pruning. In this work, we extensively evaluate our method on several architectures and tasks, including ResNe(X)t, DenseNet, MobileNet and Vision transformer for images, GAT for graph, DGCNN for 3D point cloud, alongside LSTM for language, and demonstrate that, even with a simple norm-based criterion, the proposed method consistently yields gratifying performances.

Improving the deployment efficiency of transformer-based language models has been challenging given their high computation and memory cost. While INT8 quantization has recently been shown to be effective in reducing both the memory cost and latency while preserving model accuracy, it remains unclear whether we can leverage INT4 (which doubles peak hardware throughput) to achieve further latency improvement. In this study, we explore the feasibility of employing INT4 weight and activation (W4A4) quantization for language models. Our findings indicate that W4A4 quantization introduces no to negligible accuracy degradation for encoder-only and encoder-decoder models, but causes a significant accuracy drop for decoder-only models. To materialize the performance gain using W4A4, we develop a highly optimized end-to-end W4A4 encoder inference pipeline supporting different quantization strategies. Our INT4 pipeline is 8.5times faster for latency-oriented scenarios and up to 3times for throughput-oriented scenarios compared to the inference of FP16, and improves the SOTA BERT INT8 performance from FasterTransformer by up to 1.7times. We provide insights into the failure cases when applying W4A4 to decoder-only models, and further explore the compatibility of INT4 quantization with other compression methods, like pruning and layer reduction.

Large Language Models (LLMs) have revolutionized the field of natural language processing, achieving unprecedented performance across a variety of applications by leveraging increased model sizes and sequence lengths. However, the associated rise in computational and memory costs poses significant challenges, particularly in managing long sequences due to the quadratic complexity of the transformer attention mechanism. This paper focuses on the long-context scenario, addressing the inefficiencies in KV cache memory consumption during inference. Unlike existing approaches that optimize the memory based on the sequence lengths, we uncover that the channel dimension of the KV cache exhibits significant redundancy, characterized by unbalanced magnitude distribution and low-rank structure in attention weights. Based on these observations, we propose ThinK, a novel query-dependent KV cache pruning method designed to minimize attention weight loss while selectively pruning the least significant channels. Our approach not only maintains or enhances model accuracy but also achieves a reduction in memory costs by over 20% compared with vanilla KV cache eviction methods. Extensive evaluations on the LLaMA3 and Mistral models across various long-sequence datasets confirm the efficacy of ThinK, setting a new precedent for efficient LLM deployment without compromising performance. We also outline the potential of extending our method to value cache pruning, demonstrating ThinK's versatility and broad applicability in reducing both memory and computational overheads.

As inference-time scaling becomes critical for enhanced reasoning capabilities, it is increasingly becoming important to build models that are efficient to infer. We introduce Nemotron-H, a family of 8B and 56B/47B hybrid Mamba-Transformer models designed to reduce inference cost for a given accuracy level. To achieve this goal, we replace the majority of self-attention layers in the common Transformer model architecture with Mamba layers that perform constant computation and require constant memory per generated token. We show that Nemotron-H models offer either better or on-par accuracy compared to other similarly-sized state-of-the-art open-sourced Transformer models (e.g., Qwen-2.5-7B/72B and Llama-3.1-8B/70B), while being up to 3times faster at inference. To further increase inference speed and reduce the memory required at inference time, we created Nemotron-H-47B-Base from the 56B model using a new compression via pruning and distillation technique called MiniPuzzle. Nemotron-H-47B-Base achieves similar accuracy to the 56B model, but is 20% faster to infer. In addition, we introduce an FP8-based training recipe and show that it can achieve on par results with BF16-based training. This recipe is used to train the 56B model. All Nemotron-H models will be released, with support in Hugging Face, NeMo, and Megatron-LM.

Scaling up model size and training data has advanced foundation models for instance-level perception, achieving state-of-the-art in-domain and zero-shot performance across object detection and segmentation. However, their high computational cost limits adoption on resource-constrained platforms. We first examine the limitations of existing architectures in enabling efficient edge deployment without compromising performance. We then introduce MOBIUS, a family of foundation models for universal instance segmentation, designed for Pareto-optimal downscaling to support deployment across devices ranging from high-end accelerators to mobile hardware. To reduce training and inference demands, we propose: (i) a bottleneck pixel decoder for efficient multi-scale and multi-modal fusion, (ii) a language-guided uncertainty calibration loss for adaptive decoder pruning, and (iii) a streamlined, unified training strategy. Unlike efficient baselines that trade accuracy for reduced complexity, MOBIUS reduces pixel and transformer decoder FLOPs by up to 55% and 75%, respectively, while maintaining state-of-the-art performance in just a third of the training iterations. MOBIUS establishes a new benchmark for efficient segmentation on both high-performance computing platforms and mobile devices.

Recent Vision Transformer Compression (VTC) works mainly follow a two-stage scheme, where the importance score of each model unit is first evaluated or preset in each submodule, followed by the sparsity score evaluation according to the target sparsity constraint. Such a separate evaluation process induces the gap between importance and sparsity score distributions, thus causing high search costs for VTC. In this work, for the first time, we investigate how to integrate the evaluations of importance and sparsity scores into a single stage, searching the optimal subnets in an efficient manner. Specifically, we present OFB, a cost-efficient approach that simultaneously evaluates both importance and sparsity scores, termed Once for Both (OFB), for VTC. First, a bi-mask scheme is developed by entangling the importance score and the differentiable sparsity score to jointly determine the pruning potential (prunability) of each unit. Such a bi-mask search strategy is further used together with a proposed adaptive one-hot loss to realize the progressive-and-efficient search for the most important subnet. Finally, Progressive Masked Image Modeling (PMIM) is proposed to regularize the feature space to be more representative during the search process, which may be degraded by the dimension reduction. Extensive experiments demonstrate that OFB can achieve superior compression performance over state-of-the-art searching-based and pruning-based methods under various Vision Transformer architectures, meanwhile promoting search efficiency significantly, e.g., costing one GPU search day for the compression of DeiT-S on ImageNet-1K.

Recently, state-of-the-art approaches for pruning large pre-trained models (LPMs) have demonstrated that the training-free removal of non-critical residual blocks in Transformers is viable for reducing model size, achieving results that outperform previous training-free pruning approaches. Motivated by these findings, we extend BlockPruner (Zhong et al., 2024) and propose MultiPruner, a pruning approach that surpasses recent training-free pruning methods by adopting a multidimensional, iterative, fine-grained pruning strategy. In MultiPruner, multidimensional pruning reinstates the structural balance in block-pruned models by sequentially compressing along three dimensions: i) residual blocks, ii) channels of multilayer perceptrons (MLP), and iii) attention heads. This solution enhances zero-shot accuracy on downstream tasks compared to other techniques while improving model compression ratios, producing compressed models with fewer computing and memory requirements. Extensive experiments demonstrate the advantages of the proposed method across various large pre-trained models. The code and pruning configurations are available at https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning.

Deep learning models have become increasingly popular for a wide range of applications, including computer vision, natural language processing, and speech recognition. However, these models typically require large amounts of computational resources, making them challenging to run on low-power devices such as the Raspberry Pi. One approach to addressing this challenge is to use pruning techniques to reduce the size of the deep learning models. Pruning involves removing unimportant weights and connections from the model, resulting in a smaller and more efficient model. Pruning can be done during training or after the model has been trained. Another approach is to optimize the deep learning models specifically for the Raspberry Pi architecture. This can include optimizing the model's architecture and parameters to take advantage of the Raspberry Pi's hardware capabilities, such as its CPU and GPU. Additionally, the model can be optimized for energy efficiency by minimizing the amount of computation required. Pruning and optimizing deep learning models for the Raspberry Pi can help overcome the computational and energy constraints of low-power devices, making it possible to run deep learning models on a wider range of devices. In the following sections, we will explore these approaches in more detail and discuss their effectiveness for optimizing deep learning models for the Raspberry Pi.

Pruning assumes a subnetwork exists in the original deep neural network, which can achieve comparative model performance with less computation than the original. However, it is unclear how the model performance varies with the different subnetwork extractions. In this paper, we choose the representation dimension (or embedding dimension, model dimension, the dimension of the residual stream in the relevant literature) as the entry point to this issue. We investigate the linear transformations in the LLM transformer blocks and consider a specific structured pruning approach, SliceGPT, to extract the subnetworks of different representation dimensions. We mechanistically analyse the activation flow during the model forward passes, and find the representation dimension dominates the linear transformations, model predictions, and, finally, the model performance. Explicit analytical relations are given to calculate the pruned model performance (perplexity and accuracy) without actual evaluation, and are empirically validated with Llama-3-8B-Instruct and Phi-3-mini-4k-Instruct.

Transformers have greatly advanced the state-of-the-art in Natural Language Processing (NLP) in recent years, but present very large computation and storage requirements. We observe that the design process of Transformers (pre-train a foundation model on a large dataset in a self-supervised manner, and subsequently fine-tune it for different downstream tasks) leads to task-specific models that are highly over-parameterized, adversely impacting both accuracy and inference efficiency. We propose AxFormer, a systematic framework that applies accuracy-driven approximations to create optimized transformer models for a given downstream task. AxFormer combines two key optimizations -- accuracy-driven pruning and selective hard attention. Accuracy-driven pruning identifies and removes parts of the fine-tuned transformer that hinder performance on the given downstream task. Sparse hard-attention optimizes attention blocks in selected layers by eliminating irrelevant word aggregations, thereby helping the model focus only on the relevant parts of the input. In effect, AxFormer leads to models that are more accurate, while also being faster and smaller. Our experiments on GLUE and SQUAD tasks show that AxFormer models are up to 4.5% more accurate, while also being up to 2.5X faster and up to 3.2X smaller than conventional fine-tuned models. In addition, we demonstrate that AxFormer can be combined with previous efforts such as distillation or quantization to achieve further efficiency gains.

Transformer-based language models are applied to a wide range of applications in natural language processing. However, they are inefficient and difficult to deploy. In recent years, many compression algorithms have been proposed to increase the implementation efficiency of large Transformer-based models on target hardware. In this work we present a new method for training sparse pre-trained Transformer language models by integrating weight pruning and model distillation. These sparse pre-trained models can be used to transfer learning for a wide range of tasks while maintaining their sparsity pattern. We demonstrate our method with three known architectures to create sparse pre-trained BERT-Base, BERT-Large and DistilBERT. We show how the compressed sparse pre-trained models we trained transfer their knowledge to five different downstream natural language tasks with minimal accuracy loss. Moreover, we show how to further compress the sparse models' weights to 8bit precision using quantization-aware training. For example, with our sparse pre-trained BERT-Large fine-tuned on SQuADv1.1 and quantized to 8bit we achieve a compression ratio of 40X for the encoder with less than 1% accuracy loss. To the best of our knowledge, our results show the best compression-to-accuracy ratio for BERT-Base, BERT-Large, and DistilBERT.

Generative Pre-trained Transformers (GPTs) have demonstrated remarkable performance across diverse domains through the extensive scaling of model parameters. Recent works observe the redundancy across the transformer blocks and develop compression methods by structured pruning of the unimportant blocks. However, such straightforward elimination will always provide irreversible performance degradation. In this paper, we propose FuseGPT, a novel methodology to recycle the pruned transformer blocks to further recover the model performance. Firstly we introduce a new importance detection metric, Macro Influence (MI), to detect the long-term influence of each transformer block by calculating their loss of information after removal. Then we propose group-level layers fusion, which adopts the parameters in layers of the unimportant blocks and injects them into the corresponding layers inside the neighboring blocks. The fusion is not one-off but through iterative parameter updates by lightweight group-level fine-tuning. Specifically, these injected parameters are frozen but weighted with learnable rank decomposition matrices to reduce the overhead during fine-tuning. Our approach not only works well on large language models but also on large multimodal models. The experiments have shown that, by using modest amounts of data, FuseGPT can outperform previous works in both perplexity and zero-shot task performance.

In recent years, the rapid advancement of large-scale pre-trained language models based on transformer architectures has revolutionized natural language processing tasks. Among these, ChatGPT has gained widespread popularity, demonstrating human-level conversational abilities and attracting over 100 million monthly users by late 2022. Concurrently, Baidu's commercial deployment of the Ernie Wenxin model has significantly enhanced marketing effectiveness through AI-driven technologies. This paper focuses on optimizing high-performance inference for Ernie models, emphasizing GPU acceleration and leveraging the Paddle inference framework. We employ techniques such as Faster Transformer for efficient model processing, embedding layer pruning to reduce computational overhead, and FP16 half-precision inference for enhanced computational efficiency. Additionally, our approach integrates efficient data handling strategies using multi-process parallel processing to minimize latency. Experimental results demonstrate that our optimized solution achieves up to an 8.96x improvement in inference speed compared to standard methods, while maintaining competitive performance.

Transformer-based language models have become the standard approach to solving natural language processing tasks. However, industry adoption usually requires the maximum throughput to comply with certain latency constraints that prevents Transformer models from being used in production. To address this gap, model compression techniques such as quantization and pruning may be used to improve inference efficiency. However, these compression techniques require specialized software to apply and deploy at scale. In this work, we propose a new pipeline for creating and running Fast Transformer models on CPUs, utilizing hardware-aware pruning, knowledge distillation, quantization, and our own Transformer inference runtime engine with optimized kernels for sparse and quantized operators. We demonstrate the efficiency of our pipeline by creating a Fast DistilBERT model showing minimal accuracy loss on the question-answering SQuADv1.1 benchmark, and throughput results under typical production constraints and environments. Our results outperform existing state-of-the-art Neural Magic's DeepSparse runtime performance by up to 50% and up to 4.1x performance speedup over ONNX Runtime. Source code is publicly available at https://github.com/intel/intel-extension-for-transformers.

We introduce Token Merging (ToMe), a simple method to increase the throughput of existing ViT models without needing to train. ToMe gradually combines similar tokens in a transformer using a general and light-weight matching algorithm that is as fast as pruning while being more accurate. Off-the-shelf, ToMe can 2x the throughput of state-of-the-art ViT-L @ 512 and ViT-H @ 518 models on images and 2.2x the throughput of ViT-L on video with only a 0.2-0.3% accuracy drop in each case. ToMe can also easily be applied during training, improving in practice training speed up to 2x for MAE fine-tuning on video. Training with ToMe further minimizes accuracy drop, leading to 2x the throughput of ViT-B on audio for only a 0.4% mAP drop. Qualitatively, we find that ToMe merges object parts into one token, even over multiple frames of video. Overall, ToMe's accuracy and speed are competitive with state-of-the-art on images, video, and audio.

We introduce GateSkip, a simple residual-stream gating mechanism that enables token-wise layer skipping in decoder-only LMs. Each Attention/MLP branch is equipped with a sigmoid-linear gate that condenses the branch's output before it re-enters the residual stream. During inference we rank tokens by the gate values and skip low-importance ones using a per-layer budget. While early-exit or router-based Mixture-of-Depths models are known to be unstable and need extensive retraining, our smooth, differentiable gates fine-tune stably on top of pretrained models. On long-form reasoning, we save up to 15\% compute while retaining over 90\% of baseline accuracy. On instruction-tuned models we see accuracy gains at full compute and match baseline quality near 50\% savings. The learned gates give insight into transformer information flow (e.g., BOS tokens act as anchors), and the method combines easily with quantization, pruning, and self-speculative decoding.

As the size of transformer-based models continues to grow, fine-tuning these large-scale pretrained vision models for new tasks has become increasingly parameter-intensive. Parameter-efficient learning has been developed to reduce the number of tunable parameters during fine-tuning. Although these methods show promising results, there is still a significant performance gap compared to full fine-tuning. To address this challenge, we propose an Effective and Efficient Visual Prompt Tuning (E^2VPT) approach for large-scale transformer-based model adaptation. Specifically, we introduce a set of learnable key-value prompts and visual prompts into self-attention and input layers, respectively, to improve the effectiveness of model fine-tuning. Moreover, we design a prompt pruning procedure to systematically prune low importance prompts while preserving model performance, which largely enhances the model's efficiency. Empirical results demonstrate that our approach outperforms several state-of-the-art baselines on two benchmarks, with considerably low parameter usage (e.g., 0.32% of model parameters on VTAB-1k). Our code is available at https://github.com/ChengHan111/E2VPT.

We propose Confidence-Guided Token Merging (Co-Me), an acceleration mechanism for visual geometric transformers without retraining or finetuning the base model. Co-Me distilled a light-weight confidence predictor to rank tokens by uncertainty and selectively merge low-confidence ones, effectively reducing computation while maintaining spatial coverage. Compared to similarity-based merging or pruning, the confidence signal in Co-Me reliably indicates regions emphasized by the transformer, enabling substantial acceleration without degrading performance. Co-Me applies seamlessly to various multi-view and streaming visual geometric transformers, achieving speedups that scale with sequence length. When applied to VGGT and MapAnything, Co-Me achieves up to 11.3times and 7.2times speedup, making visual geometric transformers practical for real-time 3D perception and reconstruction.

Transformers have been successfully applied in the field of video-based 3D human pose estimation. However, the high computational costs of these video pose transformers (VPTs) make them impractical on resource-constrained devices. In this paper, we present a hierarchical plug-and-play pruning-and-recovering framework, called Hierarchical Hourglass Tokenizer (H_{2}OT), for efficient transformer-based 3D human pose estimation from videos. H_{2}OT begins with progressively pruning pose tokens of redundant frames and ends with recovering full-length sequences, resulting in a few pose tokens in the intermediate transformer blocks and thus improving the model efficiency. It works with two key modules, namely, a Token Pruning Module (TPM) and a Token Recovering Module (TRM). TPM dynamically selects a few representative tokens to eliminate the redundancy of video frames, while TRM restores the detailed spatio-temporal information based on the selected tokens, thereby expanding the network output to the original full-length temporal resolution for fast inference. Our method is general-purpose: it can be easily incorporated into common VPT models on both seq2seq and seq2frame pipelines while effectively accommodating different token pruning and recovery strategies. In addition, our H_{2}OT reveals that maintaining the full pose sequence is unnecessary, and a few pose tokens of representative frames can achieve both high efficiency and estimation accuracy. Extensive experiments on multiple benchmark datasets demonstrate both the effectiveness and efficiency of the proposed method. Code and models are available at https://github.com/NationalGAILab/HoT.

Model compression offers a promising path to reducing the cost and inaccessibility of large pre-trained models, without significantly compromising their impressive performance. Large Transformer models, including large language models (LLMs), often contain computational redundancy, which can serve as a target for new model compression methods. In this work, we specifically target neuron-level redundancies in model layers by combining groups of similar neurons into fewer neurons. We frame this width reduction as a Discrete Optimal Transport problem, and propose DOTResize, a novel Transformer compression method that uses optimal transport theory to transform and compress model weights. To ensure applicability within the Transformer architecture, we motivate and incorporate entropic regularization and matrix factorization into the transportation maps produced by our method. Unlike pruning-based approaches which discard neurons based on importance measures, DOTResize re-projects the entire neuron width, allowing the retention and redistribution of useful signal across the reduced layer. Empirical results show that compared to simple or state-of-the-art neuron width-pruning techniques, DOTResize can outperform these methods across multiple LLM families and sizes, while achieving measurable reductions in real-world computational cost.

The Vision Transformer architecture is a deep learning model inspired by the success of the Transformer model in Natural Language Processing. However, the self-attention mechanism, large number of parameters, and the requirement for a substantial amount of training data still make Vision Transformers computationally burdensome. In this research, we investigate the possibility of applying Sparse Regularization to Vision Transformers and the impact of Pruning, either after Sparse Regularization or without it, on the trade-off between performance and efficiency. To accomplish this, we apply Sparse Regularization and Pruning methods to the Vision Transformer architecture for image classification tasks on the CIFAR-10, CIFAR-100, and ImageNet-100 datasets. The training process for the Vision Transformer model consists of two parts: pre-training and fine-tuning. Pre-training utilizes ImageNet21K data, followed by fine-tuning for 20 epochs. The results show that when testing with CIFAR-100 and ImageNet-100 data, models with Sparse Regularization can increase accuracy by 0.12%. Furthermore, applying pruning to models with Sparse Regularization yields even better results. Specifically, it increases the average accuracy by 0.568% on CIFAR-10 data, 1.764% on CIFAR-100, and 0.256% on ImageNet-100 data compared to pruning models without Sparse Regularization. Code can be accesed here: https://github.com/yogiprsty/Sparse-ViT

Modern deep neural networks, particularly recent large language models, come with massive model sizes that require significant computational and storage resources. To enable the deployment of modern models on resource-constrained environments and accelerate inference time, researchers have increasingly explored pruning techniques as a popular research direction in neural network compression. However, there is a dearth of up-to-date comprehensive review papers on pruning. To address this issue, in this survey, we provide a comprehensive review of existing research works on deep neural network pruning in a taxonomy of 1) universal/specific speedup, 2) when to prune, 3) how to prune, and 4) fusion of pruning and other compression techniques. We then provide a thorough comparative analysis of seven pairs of contrast settings for pruning (e.g., unstructured/structured) and explore emerging topics, including post-training pruning, different levels of supervision for pruning, and broader applications (e.g., adversarial robustness) to shed light on the commonalities and differences of existing methods and lay the foundation for further method development. To facilitate future research, we build a curated collection of datasets, networks, and evaluations on different applications. Finally, we provide some valuable recommendations on selecting pruning methods and prospect promising research directions. We build a repository at https://github.com/hrcheng1066/awesome-pruning.

We propose Long-LRM, a generalizable 3D Gaussian reconstruction model that is capable of reconstructing a large scene from a long sequence of input images. Specifically, our model can process 32 source images at 960x540 resolution within only 1.3 seconds on a single A100 80G GPU. Our architecture features a mixture of the recent Mamba2 blocks and the classical transformer blocks which allowed many more tokens to be processed than prior work, enhanced by efficient token merging and Gaussian pruning steps that balance between quality and efficiency. Unlike previous feed-forward models that are limited to processing 1~4 input images and can only reconstruct a small portion of a large scene, Long-LRM reconstructs the entire scene in a single feed-forward step. On large-scale scene datasets such as DL3DV-140 and Tanks and Temples, our method achieves performance comparable to optimization-based approaches while being two orders of magnitude more efficient. Project page: https://arthurhero.github.io/projects/llrm

Robust generalization is a major challenge in deep learning, particularly when the number of trainable parameters is very large. In general, it is very difficult to know if the network has memorized a particular set of examples or understood the underlying rule (or both). Motivated by this challenge, we study an interpretable model where generalizing representations are understood analytically, and are easily distinguishable from the memorizing ones. Namely, we consider multi-layer perceptron (MLP) and Transformer architectures trained on modular arithmetic tasks, where (xi cdot 100%) of labels are corrupted (i.e. some results of the modular operations in the training set are incorrect). We show that (i) it is possible for the network to memorize the corrupted labels and achieve 100% generalization at the same time; (ii) the memorizing neurons can be identified and pruned, lowering the accuracy on corrupted data and improving the accuracy on uncorrupted data; (iii) regularization methods such as weight decay, dropout and BatchNorm force the network to ignore the corrupted data during optimization, and achieve 100% accuracy on the uncorrupted dataset; and (iv) the effect of these regularization methods is (``mechanistically'') interpretable: weight decay and dropout force all the neurons to learn generalizing representations, while BatchNorm de-amplifies the output of memorizing neurons and amplifies the output of the generalizing ones. Finally, we show that in the presence of regularization, the training dynamics involves two consecutive stages: first, the network undergoes grokking dynamics reaching high train and test accuracy; second, it unlearns the memorizing representations, where the train accuracy suddenly jumps from 100% to 100 (1-xi)%.

Prompt-tuning methods have shown comparable performance as parameter-efficient fine-tuning (PEFT) methods in various natural language understanding tasks. However, existing prompt tuning methods still utilize the entire model architecture; thus, they fail to accelerate inference speed in the application. In this paper, we propose a novel approach called SKIll-localized Prompt tuning (SKIP), which is extremely efficient in inference time. Our method significantly enhances inference efficiency by investigating and utilizing a skill-localized subnetwork in a language model. Surprisingly, our method improves the inference speed up to 160% while pruning 52% of the parameters. Furthermore, we demonstrate that our method is applicable across various transformer-based architectures, thereby confirming its practicality and scalability.

Self-supervised speech representation learning has shown promising results in various speech processing tasks. However, the pre-trained models, e.g., HuBERT, are storage-intensive Transformers, limiting their scope of applications under low-resource settings. To this end, we propose LightHuBERT, a once-for-all Transformer compression framework, to find the desired architectures automatically by pruning structured parameters. More precisely, we create a Transformer-based supernet that is nested with thousands of weight-sharing subnets and design a two-stage distillation strategy to leverage the contextualized latent representations from HuBERT. Experiments on automatic speech recognition (ASR) and the SUPERB benchmark show the proposed LightHuBERT enables over 10^9 architectures concerning the embedding dimension, attention dimension, head number, feed-forward network ratio, and network depth. LightHuBERT outperforms the original HuBERT on ASR and five SUPERB tasks with the HuBERT size, achieves comparable performance to the teacher model in most tasks with a reduction of 29% parameters, and obtains a 3.5times compression ratio in three SUPERB tasks, e.g., automatic speaker verification, keyword spotting, and intent classification, with a slight accuracy loss. The code and pre-trained models are available at https://github.com/mechanicalsea/lighthubert.

Lightweight inference is critical for biomolecular structure prediction and other downstream tasks, enabling efficient real-world deployment and inference-time scaling for large-scale applications. In this work, we address the challenge of balancing model efficiency and prediction accuracy by making several key modifications, 1) Multi-step AF3 sampler is replaced by a few-step ODE sampler, significantly reducing computational overhead for the diffusion module part during inference; 2) In the open-source Protenix framework, a subset of pairformer or diffusion transformer blocks doesn't make contributions to the final structure prediction, presenting opportunities for architectural pruning and lightweight redesign; 3) A model incorporating an ESM module is trained to substitute the conventional MSA module, reducing MSA preprocessing time. Building on these key insights, we present Protenix-Mini, a compact and optimized model designed for efficient protein structure prediction. This streamlined version incorporates a more efficient architectural design with a two-step Ordinary Differential Equation (ODE) sampling strategy. By eliminating redundant Transformer components and refining the sampling process, Protenix-Mini significantly reduces model complexity with slight accuracy drop. Evaluations on benchmark datasets demonstrate that it achieves high-fidelity predictions, with only a negligible 1 to 5 percent decrease in performance on benchmark datasets compared to its full-scale counterpart. This makes Protenix-Mini an ideal choice for applications where computational resources are limited but accurate structure prediction remains crucial.

Model pruning is a popular approach to enable the deployment of large deep learning models on edge devices with restricted computational or storage capacities. Although sparse models achieve performance comparable to that of their dense counterparts at the level of the entire dataset, they exhibit high accuracy drops for some data sub-groups. Existing methods to mitigate this disparate impact induced by pruning (i) rely on surrogate metrics that address the problem indirectly and have limited interpretability; or (ii) scale poorly with the number of protected sub-groups in terms of computational cost. We propose a constrained optimization approach that directly addresses the disparate impact of pruning: our formulation bounds the accuracy change between the dense and sparse models, for each sub-group. This choice of constraints provides an interpretable success criterion to determine if a pruned model achieves acceptable disparity levels. Experimental results demonstrate that our technique scales reliably to problems involving large models and hundreds of protected sub-groups.

Most existing pruning works are resource-intensive, requiring retraining or fine-tuning of the pruned models for accuracy. We propose a retraining-free pruning method based on hyperspherical learning and loss penalty terms. The proposed loss penalty term pushes some of the model weights far from zero, while the rest weight values are pushed near zero and can be safely pruned with no need for retraining and a negligible accuracy drop. In addition, our proposed method can instantly recover the accuracy of a pruned model by replacing the pruned values with their mean value. Our method obtains state-of-the-art results in retraining-free pruning and is evaluated on ResNet-18/50 and MobileNetV2 with ImageNet dataset. One can easily get a 50\% pruned ResNet18 model with a 0.47\% accuracy drop. With fine-tuning, the experiment results show that our method can significantly boost the accuracy of the pruned models compared with existing works. For example, the accuracy of a 70\% pruned (except the first convolutional layer) MobileNetV2 model only drops 3.5\%, much less than the 7\% sim 10\% accuracy drop with conventional methods.

Pruning is a promising approach to compress complex deep learning models in order to deploy them on resource-constrained edge devices. However, many existing pruning solutions are based on unstructured pruning, which yields models that cannot efficiently run on commodity hardware and require users to manually explore and tune the pruning process, which is time-consuming and often leads to sub-optimal results. To address these limitations, this paper presents an adaptive, activation-based, structured pruning approach to automatically and efficiently generate small, accurate, and hardware-efficient models that meet user requirements. First, it proposes iterative structured pruning using activation-based attention feature maps to effectively identify and prune unimportant filters. Then, it proposes adaptive pruning policies for automatically meeting the pruning objectives of accuracy-critical, memory-constrained, and latency-sensitive tasks. A comprehensive evaluation shows that the proposed method can substantially outperform the state-of-the-art structured pruning works on CIFAR-10 and ImageNet datasets. For example, on ResNet-56 with CIFAR-10, without any accuracy drop, our method achieves the largest parameter reduction (79.11%), outperforming the related works by 22.81% to 66.07%, and the largest FLOPs reduction (70.13%), outperforming the related works by 14.13% to 26.53%.

Model pruning seeks to induce sparsity in a deep neural network's various connection matrices, thereby reducing the number of nonzero-valued parameters in the model. Recent reports (Han et al., 2015; Narang et al., 2017) prune deep networks at the cost of only a marginal loss in accuracy and achieve a sizable reduction in model size. This hints at the possibility that the baseline models in these experiments are perhaps severely over-parameterized at the outset and a viable alternative for model compression might be to simply reduce the number of hidden units while maintaining the model's dense connection structure, exposing a similar trade-off in model size and accuracy. We investigate these two distinct paths for model compression within the context of energy-efficient inference in resource-constrained environments and propose a new gradual pruning technique that is simple and straightforward to apply across a variety of models/datasets with minimal tuning and can be seamlessly incorporated within the training process. We compare the accuracy of large, but pruned models (large-sparse) and their smaller, but dense (small-dense) counterparts with identical memory footprint. Across a broad range of neural network architectures (deep CNNs, stacked LSTM, and seq2seq LSTM models), we find large-sparse models to consistently outperform small-dense models and achieve up to 10x reduction in number of non-zero parameters with minimal loss in accuracy.

Autoregressive Transformers adopted in Large Language Models (LLMs) are hard to scale to long sequences. Despite several works trying to reduce their computational cost, most of LLMs still adopt attention layers between all pairs of tokens in the sequence, thus incurring a quadratic cost. In this study, we present a novel approach that dynamically prunes contextual information while preserving the model's expressiveness, resulting in reduced memory and computational requirements during inference. Our method employs a learnable mechanism that determines which uninformative tokens can be dropped from the context at any point across the generation process. By doing so, our approach not only addresses performance concerns but also enhances interpretability, providing valuable insight into the model's decision-making process. Our technique can be applied to existing pre-trained models through a straightforward fine-tuning process, and the pruning strength can be specified by a sparsity parameter. Notably, our empirical findings demonstrate that we can effectively prune up to 80\% of the context without significant performance degradation on downstream tasks, offering a valuable tool for mitigating inference costs. Our reference implementation achieves up to 2times increase in inference throughput and even greater memory savings.

We present a comprehensive report on compressing the Llama 3.1 8B and Mistral NeMo 12B models to 4B and 8B parameters, respectively, using pruning and distillation. We explore two distinct pruning strategies: (1) depth pruning and (2) joint hidden/attention/MLP (width) pruning, and evaluate the results on common benchmarks from the LM Evaluation Harness. The models are then aligned with NeMo Aligner and tested in instruct-tuned versions. This approach produces a compelling 4B model from Llama 3.1 8B and a state-of-the-art Mistral-NeMo-Minitron-8B (MN-Minitron-8B for brevity) model from Mistral NeMo 12B. We found that with no access to the original data, it is beneficial to slightly fine-tune teacher models on the distillation dataset. We open-source our base model weights on Hugging Face with a permissive license.

In the present work we present Training Noise Token (TNT) Pruning for vision transformers. Our method relaxes the discrete token dropping condition to continuous additive noise, providing smooth optimization in training, while retaining discrete dropping computational gains in deployment settings. We provide theoretical connections to Rate-Distortion literature, and empirical evaluations on the ImageNet dataset using ViT and DeiT architectures demonstrating TNT's advantages over previous pruning methods.

Since its inception, Vision Transformer (ViT) has emerged as a prevalent model in the computer vision domain. Nonetheless, the multi-head self-attention (MHSA) mechanism in ViT is computationally expensive due to its calculation of relationships among all tokens. Although some techniques mitigate computational overhead by discarding tokens, this also results in the loss of potential information from those tokens. To tackle these issues, we propose a novel token pruning method that retains information from non-crucial tokens by merging them with more crucial tokens, thereby mitigating the impact of pruning on model performance. Crucial and non-crucial tokens are identified by their importance scores and merged based on similarity scores. Furthermore, multi-scale features are exploited to represent images, which are fused prior to token pruning to produce richer feature representations. Importantly, our method can be seamlessly integrated with various ViTs, enhancing their adaptability. Experimental evidence substantiates the efficacy of our approach in reducing the influence of token pruning on model performance. For instance, on the ImageNet dataset, it achieves a remarkable 33% reduction in computational costs while only incurring a 0.1% decrease in accuracy on DeiT-S.

In recent years, deep network pruning has attracted significant attention in order to enable the rapid deployment of AI into small devices with computation and memory constraints. Pruning is often achieved by dropping redundant weights, neurons, or layers of a deep network while attempting to retain a comparable test performance. Many deep pruning algorithms have been proposed with impressive empirical success. However, existing approaches lack a quantifiable measure to estimate the compressibility of a sub-network during each pruning iteration and thus may under-prune or over-prune the model. In this work, we propose PQ Index (PQI) to measure the potential compressibility of deep neural networks and use this to develop a Sparsity-informed Adaptive Pruning (SAP) algorithm. Our extensive experiments corroborate the hypothesis that for a generic pruning procedure, PQI decreases first when a large model is being effectively regularized and then increases when its compressibility reaches a limit that appears to correspond to the beginning of underfitting. Subsequently, PQI decreases again when the model collapse and significant deterioration in the performance of the model start to occur. Additionally, our experiments demonstrate that the proposed adaptive pruning algorithm with proper choice of hyper-parameters is superior to the iterative pruning algorithms such as the lottery ticket-based pruning methods, in terms of both compression efficiency and robustness.

The adoption of Vision Transformers (ViTs) in resource-constrained applications necessitates improvements in inference throughput. To this end several token pruning and merging approaches have been proposed that improve efficiency by successively reducing the number of tokens. However, it remains an open problem to design a token reduction method that is fast, maintains high performance, and is applicable to various vision tasks. In this work, we present a token pruner that uses auxiliary prediction heads that learn to select tokens end-to-end based on task relevance. These auxiliary heads can be removed after training, leading to throughput close to that of a random pruner. We evaluate our method on image classification, semantic segmentation, object detection, and instance segmentation, and show speedups of 1.5 to 4x with small drops in performance. As a best case, on the ADE20k semantic segmentation benchmark, we observe a 2x speedup relative to the no-pruning baseline, with a negligible performance penalty of 0.1 median mIoU across 5 seeds.

Pruning has become a widely adopted technique for reducing the hardware requirements of large language models (LLMs). To recover model performance after pruning, post-training is commonly employed to mitigate the resulting performance degradation. While post-training benefits from larger datasets, once the dataset size is already substantial, increasing the training data provides only limited performance gains. To balance post-training cost and model performance, it is necessary to explore the optimal amount of post-training data.Through extensive experiments on the Llama-3 and Qwen-2.5 series models, pruned using various common pruning methods, we uncover the scaling Law for Post-training after model Pruning, referred to as the P^2 Law.This law identifies four key factors for predicting the pruned model's post-training loss: the model size before pruning, the number of post-training tokens, the pruning rate, and the model's loss before pruning. Moreover, P^2 Law can generalize to larger dataset sizes, larger model sizes, and higher pruning rates, offering valuable insights for the post-training of pruned LLMs.

Adversarial pruning compresses models while preserving robustness. Current methods require access to adversarial examples during pruning. This significantly hampers training efficiency. Moreover, as new adversarial attacks and training methods develop at a rapid rate, adversarial pruning methods need to be modified accordingly to keep up. In this work, we propose a novel framework to prune a previously trained robust neural network while maintaining adversarial robustness, without further generating adversarial examples. We leverage concurrent self-distillation and pruning to preserve knowledge in the original model as well as regularizing the pruned model via the Hilbert-Schmidt Information Bottleneck. We comprehensively evaluate our proposed framework and show its superior performance in terms of both adversarial robustness and efficiency when pruning architectures trained on the MNIST, CIFAR-10, and CIFAR-100 datasets against five state-of-the-art attacks. Code is available at https://github.com/neu-spiral/PwoA/.

Small language models (SLMs) have attracted considerable attention from both academia and industry due to their broad range of applications in edge devices. To obtain SLMs with strong performance, conventional approaches either pre-train the models from scratch, which incurs substantial computational costs, or compress/prune existing large language models (LLMs), which results in performance drops and falls short in comparison to pre-training. In this paper, we investigate the family of acceleration methods that involve both structured pruning and model training. We found 1) layer-wise adaptive pruning (Adapt-Pruner) is extremely effective in LLMs and yields significant improvements over existing pruning techniques, 2) adaptive pruning equipped with further training leads to models comparable to those pre-training from scratch, 3) incremental pruning brings non-trivial performance gain by interleaving pruning with training and only removing a small portion of neurons (sim5%) at a time. Experimental results on LLaMA-3.1-8B demonstrate that Adapt-Pruner outperforms conventional pruning methods, such as LLM-Pruner, FLAP, and SliceGPT, by an average of 1%-7% in accuracy on commonsense benchmarks. Additionally, Adapt-Pruner restores the performance of MobileLLM-125M to 600M on the MMLU benchmark with 200times fewer tokens via pruning from its larger counterparts, and discovers a new 1B model that surpasses LLaMA-3.2-1B in multiple benchmarks.

Pruning is a popular technique for reducing the model size and computational cost of convolutional neural networks (CNNs). However, a slow retraining or fine-tuning procedure is often required to recover the accuracy loss caused by pruning. Recently, a new research direction on weight pruning, pruning-at-initialization (PAI), is proposed to directly prune CNNs before training so that fine-tuning or retraining can be avoided. While PAI has shown promising results in reducing the model size, existing approaches rely on fine-grained weight pruning which requires unstructured sparse matrix computation, making it difficult to achieve real speedup in practice unless the sparsity is very high. This work is the first to show that fine-grained weight pruning is in fact not necessary for PAI. Instead, the layerwise compression ratio is the main critical factor to determine the accuracy of a CNN model pruned at initialization. Based on this key observation, we propose PreCropping, a structured hardware-efficient model compression scheme. PreCropping directly compresses the model at the channel level following the layerwise compression ratio. Compared to weight pruning, the proposed scheme is regular and dense in both storage and computation without sacrificing accuracy. In addition, since PreCropping compresses CNNs at initialization, the computational and memory costs of CNNs are reduced for both training and inference on commodity hardware. We empirically demonstrate our approaches on several modern CNN architectures, including ResNet, ShuffleNet, and MobileNet for both CIFAR-10 and ImageNet.

Large language models (LLMs) demonstrate strong performance as text embedding models when finetuned with supervised contrastive training. However, their large size balloons inference time and memory requirements. In this paper, we show that by pruning the last p% layers of an LLM before supervised training for only 1000 steps, we can achieve a proportional reduction in memory and inference time. We evaluate four different state-of-the-art LLMs on text embedding tasks and find that our method can prune up to 30\% of layers with negligible impact on performance and up to 80\% with only a modest drop. With only three lines of code, our method is easily implemented in any pipeline for transforming LLMs to text encoders. We also propose L^3 Prune, a novel layer-pruning strategy based on the model's initial loss that provides two optimal pruning configurations: a large variant with negligible performance loss and a small variant for resource-constrained settings. On average, the large variant prunes 21\% of the parameters with a -0.3 performance drop, and the small variant only suffers from a -5.1 decrease while pruning 74\% of the model. We consider these results strong evidence that LLMs are overparameterized for text embedding tasks, and can be easily pruned.

The increasing complexity and parameter count of Convolutional Neural Networks (CNNs) and Transformers pose challenges in terms of computational efficiency and resource demands. Pruning has been identified as an effective strategy to address these challenges by removing redundant elements such as neurons, channels, or connections, thereby enhancing computational efficiency without heavily compromising performance. This paper builds on the foundational work of Optimal Brain Damage (OBD) by advancing the methodology of parameter importance estimation using the Hessian matrix. Unlike previous approaches that rely on approximations, we introduce Optimal Brain Apoptosis (OBA), a novel pruning method that calculates the Hessian-vector product value directly for each parameter. By decomposing the Hessian matrix across network layers and identifying conditions under which inter-layer Hessian submatrices are non-zero, we propose a highly efficient technique for computing the second-order Taylor expansion of parameters. This approach allows for a more precise pruning process, particularly in the context of CNNs and Transformers, as validated in our experiments including VGG19, ResNet32, ResNet50, and ViT-B/16 on CIFAR10, CIFAR100 and Imagenet datasets. Our code is available at https://github.com/NEU-REAL/OBA.

Magnitude pruning is a widely used strategy for reducing model size in pure supervised learning; however, it is less effective in the transfer learning regime that has become standard for state-of-the-art natural language processing applications. We propose the use of movement pruning, a simple, deterministic first-order weight pruning method that is more adaptive to pretrained model fine-tuning. We give mathematical foundations to the method and compare it to existing zeroth- and first-order pruning methods. Experiments show that when pruning large pretrained language models, movement pruning shows significant improvements in high-sparsity regimes. When combined with distillation, the approach achieves minimal accuracy loss with down to only 3% of the model parameters.

The ever-increasing size of large language models (LLMs) presents significant challenges for deployment due to their heavy computational and memory requirements. Current model pruning techniques attempt to alleviate these issues by relying heavily on external calibration datasets to determine which parameters to prune or compress, thus limiting their flexibility and scalability across different compression ratios. Moreover, these methods often cause severe performance degradation, particularly in downstream tasks, when subjected to higher compression rates. In this paper, we propose PruneNet, a novel model compression method that addresses these limitations by reformulating model pruning as a policy learning process. PruneNet decouples the pruning process from the model architecture, eliminating the need for calibration datasets. It learns a stochastic pruning policy to assess parameter importance solely based on intrinsic model properties while preserving the spectral structure to minimize information loss. PruneNet can compress the LLaMA-2-7B model in just 15 minutes, achieving over 80% retention of its zero-shot performance with a 30% compression ratio, outperforming existing methods that retain only 75% performance. Furthermore, on complex multitask language understanding tasks, PruneNet demonstrates its robustness by preserving up to 80% performance of the original model, proving itself a superior alternative to conventional structured compression techniques.

Pruning is a model compression method that removes redundant parameters in deep neural networks (DNNs) while maintaining accuracy. Most available filter pruning methods require complex treatments such as iterative pruning, features statistics/ranking, or additional optimization designs in the training process. In this paper, we propose a simple and effective regularization strategy from a new perspective of evolution of features, which we call feature flow regularization (FFR), for improving structured sparsity and filter pruning in DNNs. Specifically, FFR imposes controls on the gradient and curvature of feature flow along the neural network, which implicitly increases the sparsity of the parameters. The principle behind FFR is that coherent and smooth evolution of features will lead to an efficient network that avoids redundant parameters. The high structured sparsity obtained from FFR enables us to prune filters effectively. Experiments with VGGNets, ResNets on CIFAR-10/100, and Tiny ImageNet datasets demonstrate that FFR can significantly improve both unstructured and structured sparsity. Our pruning results in terms of reduction of parameters and FLOPs are comparable to or even better than those of state-of-the-art pruning methods.

Large language models (LLMs) have achieved significant progress across various domains, but their increasing scale results in high computational and memory costs. Recent studies have revealed that LLMs exhibit sparsity, providing the potential to reduce model size through pruning techniques. However, existing pruning methods typically follow a prune-then-finetune paradigm. Since the pruned components still contain valuable information, their direct removal often leads to irreversible performance degradation, imposing a substantial computational burden to recover performance during finetuning. In this paper, we propose a novel paradigm that first applies regularization, then prunes, and finally finetunes. Based on this paradigm, we introduce DReSS, a simple and effective Data-driven Regularized Structured Streamlining method for LLMs. By leveraging a small amount of data to regularize the components to be pruned, DReSS explicitly transfers the important information to the remaining parts of the model in advance. Compared to direct pruning, this can reduce the information loss caused by parameter removal, thereby enhancing its language modeling capabilities. Experimental results demonstrate that DReSS significantly outperforms existing pruning methods even under extreme pruning ratios, significantly reducing latency and increasing throughput.

Deep Neural Network (DNN) is powerful but computationally expensive and memory intensive, thus impeding its practical usage on resource-constrained front-end devices. DNN pruning is an approach for deep model compression, which aims at eliminating some parameters with tolerable performance degradation. In this paper, we propose a novel momentum-SGD-based optimization method to reduce the network complexity by on-the-fly pruning. Concretely, given a global compression ratio, we categorize all the parameters into two parts at each training iteration which are updated using different rules. In this way, we gradually zero out the redundant parameters, as we update them using only the ordinary weight decay but no gradients derived from the objective function. As a departure from prior methods that require heavy human works to tune the layer-wise sparsity ratios, prune by solving complicated non-differentiable problems or finetune the model after pruning, our method is characterized by 1) global compression that automatically finds the appropriate per-layer sparsity ratios; 2) end-to-end training; 3) no need for a time-consuming re-training process after pruning; and 4) superior capability to find better winning tickets which have won the initialization lottery.