发表机构
Institute of Artificial Intelligence and Robotics, Xi’an Jiaotong University; Jiangsu Cytoderm Intelligent Technology Co., Ltd.（西安交通大学人工智能与机器人研究所; 江苏希迪姆智能科技有限公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究视觉语言动作模型分块策略的边界抖动问题，提出ChunkFlow框架划分块区域，执行时用确定性重叠混合，训练时用多种损失，通过实验验证该框架在低延迟推理下能改善成功稳定性权衡。
AI 中文摘要
视觉语言动作（VLA）模型越来越多地采用分块动作头来满足实时约束，但这会引入边界抖动，连续块之间的重叠区域常产生不一致预测，降低时间连贯性和任务成功率。现有方法如推理时混合仅重新加权不匹配提议而不纠正根本错误。我们提出ChunkFlow，一种用于分块策略的感知接缝训练和执行框架，将块结构与边界执行对齐。它划分区域，执行时应用确定性重叠混合，用接缝及一阶和二阶连续性损失训练原始预测。实验表明其在低延迟推理下改善了成功稳定性权衡。
英文摘要
Vision-language action (VLA) models increasingly adopt chunked action heads to satisfy real-time constraints; however, this introduces boundary jitter: overlapping regions between consecutive chunks often yield inconsistent predictions, degrading temporal coherence and the task success rate. Existing methods, such as inference-time blending, merely reweight mismatched proposals without correcting underlying errors, leading to residual accumulation under biased or noisy histories. We propose ChunkFlow, a seam-aware training-and-execution framework for chunked policies that aligns chunk structure with boundary execution. It partitions each chunk into frozen, editable, and future zones, applies deterministic overlap blending at execution, and trains raw predictions with seam and first- and second-order continuity losses. History corruption and scheduled sampling improve robustness to executed-history errors, while an AWAC fine-tuning stage adapts the policy without removing these structural regularizers. Under mild smoothness assumptions, pre-blending seam discrepancies provably decay with increasing overlap. Experiments on CALVIN, LIBERO, and real robots show an improved success-stability trade-off with low-latency inference. Project page: https://cytoderm-ai.github.io/chunkflow.
