AtlasVLA: Persistent World-Ego State Modeling for Vision-Language-Action Models
- 类型：arxiv
- 标识：2608.06729
- 链接：https://arxiv.org/abs/2608.06729
- 主分类：multimodal
- 形态：method
- 被引：0
- 被引来源：Semantic Scholar
- S2被引：0
- 影响力被引：0
- TLDR：AtlasVLA is a novel framework that transitions from direct reactive manipulation to proactive reasoning through a persistent world-ego state and decisively outperforms multi-view baselines, yielding absolute success rate improvements of 9.4% on LIBERO-Long and 17.5% in real-world long-horizon tasks.
- 待LLM分类：否
- 标题中文：AtlasVLA：面向视觉—语言—动作模型的持久世界—自我状态建模
- TLDR中文：AtlasVLA 是一种新框架，通过持久化的世界-自我状态从直接反应式操作转向主动推理，显著优于多视角基线，在 LIBERO-Long 上取得 9.4% 的绝对成功率提升，在真实世界长周期任务中取得 17.5% 的提升。
- 来源文件：
- /inbox/tom/_candidates/2026-08-13-agent-rag-longcontext-candidates.json
- [S2 enrich]