发表机构
School of Computer Science, SJTU; SCUT; ECUST; ECNU; PolyU; CCUT; UESTC; HUST（上海交通大学计算机科学学院; 华南理工大学; 华东理工大学; 华东师范大学; 香港理工大学; 长春工业大学; 电子科技大学; 华中科技大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出TacForcing框架，通过流式动作专家结合执行时触觉反馈，并引入EATA减少时间不匹配，在模拟和真实接触操纵任务中均优于基线方法。
AI 中文摘要
接触丰富的操纵任务需要适应在一个动作周期内会发生显著变化的接触状态。然而，基于块的视觉-语言-动作模型会根据执行前收集的观测结果预测完整的动作块，导致执行过程中触觉条件过时。现有的触觉响应方法通常依赖独立的高频控制器，这会增加架构和训练的复杂性。本文提出了TacForcing，一种能有效结合执行时触觉反馈的流式动作生成框架。TacForcing不使用独立的响应式控制器，而是将标准动作专家替换为流式动作专家，根据执行过程中获取的不断变化的触觉观测结果生成动作。TacForcing还引入了执行感知触觉注意力（EATA），将触觉条件限制在接近执行的动作上，从而减少触觉采集与动作执行之间的时间不匹配。在六个模拟UniVTAC任务和三个真实世界接触丰富的操纵任务中，TacForcing分别达到了65%和69%的平均成功率，在两种设置下均优于强大的基线方法。
英文摘要
Contact-rich manipulation requires adapting to contact states that can evolve substantially within an action horizon. However, chunk-based vision-language-action models predict complete action chunks from observations collected before execution, leaving tactile conditioning stale during execution. Existing tactile-reactive approaches typically rely on separate high-frequency controllers, which increase both architectural and training complexity. In this paper, we introduce TacForcing, a streaming action-generation framework incorporating execution-time tactile feedback. TacForcing replaces the standard action expert with a streaming expert that generates action blocks sequentially while preserving intermediate states of unfinished blocks. After each block is executed, the expert resumes generation from these states using newly acquired tactile feedback. To better align tactile conditioning with action execution, we further introduce Execution-Aware Tactile Attention (EATA), which restricts direct access to each tactile update to the next block scheduled for execution. Across six UniVTAC simulation tasks and six real-world contact-rich manipulation tasks on two robot platforms, TacForcing achieves average success rates of 65% in simulation and 66% in the real world, outperforming the strongest baselines by 6 and 15 percentage points, respectively.
Comments15 pages, 9 figures, 2 tables. Revised text and expanded real-world experiments
