Jiwon Song

jiwonsong's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a paper 5 days ago

Joint Subspace Sparse Attention for Efficient Long-Context Decoding

Paper • 2610.19026 • Published 28 days ago • 12

upvoted a paper 7 days ago

When Does Better Acceptance Pay? A Controlled Study of Speculative Decoding

Paper • 2610.24993 • Published 21 days ago • 2

authored a paper 25 days ago

Joint Subspace Sparse Attention for Efficient Long-Context Decoding

Paper • 2610.19026 • Published 28 days ago • 12

submitted a paper to Daily Papers 26 days ago

Joint Subspace Sparse Attention for Efficient Long-Context Decoding

Paper • 2610.19026 • Published 28 days ago • 12

upvoted 3 papers about 1 month ago

WaveFront Decoding: Parallelized Self-Speculative Decoding for Looped Language Models

Paper • 2609.23033 • Published Sep 26 • 9

KVCMAS: Efficient KV cache Correction for Shared Context in Multi-Agent Systems

Paper • 2609.34060 • Published Sep 28 • 7

PReCache: Efficient KV Cache Sharing for Multi-LoRA Agents via Low-Rank Precomputation and Neutral Reconstruction

Paper • 2609.34054 • Published Sep 28 • 7

upvoted a paper 3 months ago

Rotation-Aligned Key Channel Pruning for Efficient Vision-Language Model Inference

Paper • 2605.19218 • Published May 19 • 2

upvoted a paper 4 months ago

Loop the Loopies!

Paper • 2607.16051 • Published Jul 17 • 53

authored a paper 4 months ago

Rotation-Aligned Key Channel Pruning for Efficient Vision-Language Model Inference

Paper • 2605.19218 • Published May 19 • 2

authored a paper 6 months ago

CompactAttention: Accelerating Chunked Prefill with Block-Union KV Selection

Paper • 2605.16839 • Published May 16 • 13

upvoted a paper 6 months ago

CompactAttention: Accelerating Chunked Prefill with Block-Union KV Selection

Paper • 2605.16839 • Published May 16 • 13

submitted a paper to Daily Papers 6 months ago

CompactAttention: Accelerating Chunked Prefill with Block-Union KV Selection

Paper • 2605.16839 • Published May 16 • 13

upvoted a paper 6 months ago

Teaching Language Models to Think in Code

Paper • 2605.07237 • Published May 11 • 32

updated a model 7 months ago

jiwonsong/SeerAttention-Gemma3-12B-AttnGates

Text Generation • Updated Apr 27 • 2

published a model 7 months ago

jiwonsong/SeerAttention-Gemma3-12B-AttnGates

Text Generation • Updated Apr 27 • 2

updated a model 7 months ago

jiwonsong/SeerAttention-Qwen3-8B-AttnGates

Text Generation • Updated Apr 6 • 9 • 1

published a model 7 months ago

jiwonsong/SeerAttention-Qwen3-8B-AttnGates

Text Generation • Updated Apr 6 • 9 • 1

upvoted 2 papers 8 months ago

Attention Residuals

Paper • 2603.15031 • Published Mar 16 • 198

LK Losses: Direct Acceptance Rate Optimization for Speculative Decoding

Paper • 2602.23881 • Published Feb 27 • 18