Improve model card: Add intro and fix GitHub link

README.md

@@ -2,6 +2,9 @@
 license: cc-by-nc-4.0
 pipeline_tag: image-text-to-text
 ---
+
+The AutoNeural-VL-1.5B model is an NPU-native vision-language model for in-car assistants, presented in the paper [AutoNeural: Co-Designing Vision-Language Models for NPU Inference](https://arxiv.org/abs/2512.02924).
+
 <p align="center">
   <a href="https://arxiv.org/abs/2512.02924"><img src="https://img.shields.io/badge/📄%20arXiv-2512.02924-b31b1b?style=for-the-badge" alt="arXiv"></a>
   <a href="https://discord.com/invite/nexa-ai"><img src="https://img.shields.io/badge/💬%20Discord-Nexa%20AI-5865F2?style=for-the-badge" alt="Discord"></a>
@@ -9,7 +12,7 @@ pipeline_tag: image-text-to-text
 </p>
 
 <p align="center">
-  <a href="https://github.com/NexaAI/nexa-sdk/edit/main/solutions/autoneural/README.md"><b>🌟 Github</b></a> |
+  <a href="https://github.com/NexaAI/nexa-sdk"><b>🌟 Github</b></a> |
   <a href="https://nexa.ai/solution/intelligent-cockpit"><b>📄 Webpage</b></a> 
 </p>
 
@@ -103,10 +106,10 @@ AutoNeural is an NPU-native vision–language model co-designed for integer-only
 
 AutoNeural follows a four-stage curriculum on large-scale multimodal data plus a proprietary automotive dataset.
 
-1. **Image–text alignment.** Freeze vision and language backbones; train only the projector on image–caption pairs to learn basic visual grounding.
-2. **General visual understanding.** Unfreeze the full model and train on broad VQA-style tasks (object/scene understanding, basic reasoning) from the Infinity-MM dataset to build strong general multimodal capability.
-3. **Instruction tuning.** Continue training on diverse instruction-following data (documents, charts, OCR, multi-turn dialogue, specialized domains) using a mixture of task weights for balanced performance.
-4. **Automotive domain finetuning.** Finetune on ~200k curated cockpit samples (AI Sentinel, Greeter, Car Finder, Safety when getting on/off) plus high-quality synthetic data, with an NPU-aware recipe that combines quantization-aware training, mixed-precision constraints, and calibration to keep post-quantization drift low on real hardware.
+1.  **Image–text alignment.** Freeze vision and language backbones; train only the projector on image–caption pairs to learn basic visual grounding.
+2.  **General visual understanding.** Unfreeze the full model and train on broad VQA-style tasks (object/scene understanding, basic reasoning) from the Infinity-MM dataset to build strong general multimodal capability.
+3.  **Instruction tuning.** Continue training on diverse instruction-following data (documents, charts, OCR, multi-turn dialogue, specialized domains) using a mixture of task weights for balanced performance.
+4.  **Automotive domain finetuning.** Finetune on ~200k curated cockpit samples (AI Sentinel, Greeter, Car Finder, Safety when getting on/off) plus high-quality synthetic data, with an NPU-aware recipe that combines quantization-aware training, mixed-precision constraints, and calibration to keep post-quantization drift low on real hardware.
 
 ---