paper-with-me

Papers

AtlasVA: Self-Evolving Visual Skill Memory for Teacher-Free VLM Agents

2026-05-18 · Pan Wang, Yihao Hu, Xiujin Liu, Jingchu Yang, Hang Wang, Zhihao Wen arxiv

Vision-language model (VLM) agents increasingly rely on memory-augmented reinforcement learning to reuse experience across long-horizon tasks, yet most existing frameworks store memory as text and depend on proprietary teacher models to summarize or refine it. This design is poorly matched to spatial decision making: geometric priors are compressed into lossy language, and sparse interaction is often supervised through delayed textual feedback rather than dense visually grounded signals. We argue that reusable experience for VLM agents should remain visually grounded. Based on this insight, we propose \textbf{AtlasVA}, a teacher-free visual skill memory framework that organizes memory into three complementary layers: spatial heatmaps, visual exemplars, and symbolic text skills. AtlasVA further evolves danger and affinity atlases directly from trajectory statistics and lightweight grid heuristics, and reuses these self-evolving atlases as potential-based shaping rewards for reinforcement learning. This unifies perception, memory, and optimization without external LLM supervision. Experiments on \textsc{Sokoban}, \textsc{FrozenLake}, 3D embodied navigation, and 3D robotic manipulation benchmarks show that AtlasVA consistently outperforms text-centric memory baselines and competitive VLM agents, with especially strong gains on spatially intensive tasks. Homepage: https://wangpan-ustc.github.io/AtlasvaWeb

📄 PDF Abstract BibTeX arXiv:2605.17933

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement LearningDecision Making

Similar Papers 제목 키워드 기반

MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents

2026-02-02 · Haozhen Zhang, Quanyu Long, Jianzhu Bao, Tao Feng 외 arxiv

Most Large Language Model (LLM) agent memory systems rely on a small set of static, hand-designed operations for extracting memory. These fixed procedures hard-code human priors about what to store and how to revise memo…

SkillOS: Learning Skill Curation for Self-Evolving Agents

2026-05-07 · Siru Ouyang, Jun Yan, Yanfei Chen, Rujun Han 외 arxiv

LLM-based agents are increasingly deployed to handle streaming tasks, yet they often remain one-off problem solvers that fail to learn from past interactions. Reusable skills distilled from experience provide a natural s…

MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and Evaluation

2026-05-26 · Huawei Lin, Peng Li, Jie Song, Fuxin Jiang 외 arxiv

Large language model (LLM) agents rely on reusable skills to solve complex tasks, but existing skill creation approaches often treat skills as isolated, static artifacts, limiting reusability, reliability, and long-term …

SkillJack: Persistent Skill Backdoors in Self-Evolving Agents

2026-08-04 · Zonghao Ying, Xiangfan Wu, Huiyu Wu, Xing Zheng 외 hf

Self-evolving agents increasingly convert interaction histories into reusable skills that persist beyond individual tasks. While prior work studies memory and retrieval poisoning, such attacks only affect agents when poi…

Experience Makes Skillful: Enabling Generalizable Medical Agent Reasoning via Self-Evolving Skill Memory

2026-06-08 · Haoran Sun, Wenjie Li, Yujie Zhang, Zekai Lin 외 arxiv

Medical agent systems are increasingly expected to support interactive clinical decision making rather than only static question answering. In such settings, effective agents must reuse prior experience across evolving c…

Question AnsweringDecision Making