AVA-VLA：利用主动视觉注意力改进视觉-语言-动作模型
AVA-VLA: Improving Vision-Language-Action models with Active Visual Attention
将部分可观测马尔可夫决策过程引入VLA策略学习，用循环状态近似历史信念，并基于此设计主动视觉注意力动态重加权当前视觉token，为视觉研究者提供了一种时序感知的视觉特征选择机制，可复用于需要历史推理的视觉决策任务。
原文依据 we introduce Active Visual Attention (AVA), which dynamically reweights visual tokens in the current observation to focus on regions most relevant given both the instruction and execution history.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论视觉-语言-动作（VLA）模型近期在具身任务中取得了显著进展，但大多数方法在每个时间步独立处理视觉观测。这种与历史无关的设计将机器人操作视为马尔可夫决策过程，而真实世界的机器人控制本质上是部分可观测的，并且需要对过去的交互进行推理。为了解决这一不匹配，我们从部分可观测马尔可夫决策过程的视角重新形式化VLA策略学习，并提出AVA-VLA框架，该框架以循环状态为条件生成动作，该循环状态作为智能体对任务历史信念的神经近似。基于此循环状态，我们引入了主动视觉注意力（AVA），它根据指令和执行历史动态重加权当前观测中的视觉token，以聚焦于最相关的区域。大量实验表明，AVA-VLA在标准机器人基准（包括LIBERO和CALVIN）上达到了最优性能，并有效迁移到真实世界双臂操作任务。这些结果证明了时间接地的主动视觉处理对于提升VLA在机器人序列决策中性能的有效性。
英文原文摘要 1266 字符
Vision-Language-Action (VLA) models have shown remarkable progress in embodied tasks recently, but most methods process visual observations independently at each timestep. This history-agnostic design treats robot manipulation as a Markov Decision Process, even though real-world robotic control is inherently partially observable and requires reasoning over past interactions. To address this mismatch, we reformulate VLA policy learning from a Partially Observable Markov Decision Process perspective and propose AVA-VLA, a framework that conditions action generation on a recurrent state that serves as a neural approximation to the agent's belief over task history. Built on this recurrent state, we introduce Active Visual Attention (AVA), which dynamically reweights visual tokens in the current observation to focus on regions most relevant given both the instruction and execution history. Extensive experiments show that AVA-VLA achieves state-of-the-art performance on standard robotic benchmarks, including LIBERO and CALVIN, and transfers effectively to real-world dual-arm manipulation tasks. These results demonstrate the effectiveness of temporally grounded active visual processing for improving VLA performance in robotic sequential decision-making.