| --- |
| license: mit |
| library_name: transformers |
| pipeline_tag: image-text-to-text |
| base_model: Qwen/Qwen2.5-VL-3B-Instruct |
| --- |
| |
| # AtlasVA: Self-Evolving Visual Skill Memory for Teacher-Free VLM Agents |
|
|
| This repository contains the model weights for **AtlasVA**, as presented in the paper [AtlasVA: Self-Evolving Visual Skill Memory for Teacher-Free VLM Agents](https://huggingface.co/papers/2605.17933). |
|
|
| **AtlasVA** is a teacher-free visual skill memory framework designed for Vision-Language Model (VLM) agents. It organizes memory into three complementary layers: spatial heatmaps, visual exemplars, and symbolic text skills. By evolving danger and affinity atlases directly from trajectory statistics, AtlasVA provides dense, coordinate-aware guidance for reinforcement learning, unifying perception, memory, and optimization without external LLM supervision. |
|
|
| - **Project Page:** [https://wangpan-ustc.github.io/AtlasvaWeb/](https://wangpan-ustc.github.io/AtlasvaWeb/) |
| - **Repository:** [https://github.com/wangpan-ustc/AtlasVA](https://github.com/wangpan-ustc/AtlasVA) |
| - **Paper:** [https://huggingface.co/papers/2605.17933](https://huggingface.co/papers/2605.17933) |
|
|
| ## Model Details |
| - **Base Model**: Qwen2.5-VL-3B-Instruct |
| - **Task**: Multimodal agentic decision making (Sokoban, FrozenLake, 3D navigation, robotic manipulation). |
| - **Memory Layers**: Spatial heatmaps, visual exemplars, and symbolic text skills. |
|
|
| ## Citation |
|
|
| ```bibtex |
| @article{wang2026atlasva, |
| title={AtlasVA: Self-Evolving Visual Skill Memory for Teacher-Free VLM Agents}, |
| author={Wang, Pan and Hu, Yihao and Liu, Xiujin and Yang, Jingchu and Wang, Hang and Wen, Zhihao}, |
| journal={arXiv preprint arXiv:2605.17933}, |
| year={2026} |
| } |
| ``` |
