FiberTune: 在视觉-语言-动作微调中保留动作纤维视觉残差
FiberTune: Preserving Action-Fiber Visual Residuals in Vision-Language-Action Fine-Tuning
AI总结：
提出FiberTune，通过在线动作探针过滤动作预测特征方向，对齐教师视觉残差并正则化有效秩，在六个仿真和实物任务中提升VLA策略性能。
摘要：
动作监督的视觉-语言-动作（VLA）策略微调能有效拟合演示，但仅约束改变预测动作的方向，导致动作等价状态下视觉结构自由坍缩。我们将此形式化为沿局部动作纤维的残差视觉坍缩，并提出FiberTune，一种训练时目标，在不增加推理开销的情况下保留教师结构的视觉残差。FiberTune使用在线动作探针估计动作预测特征方向，从中滤除中间视觉标记表示，并将探针过滤后的残差与冻结的视觉教师对齐，同时正则化其有效秩。在相同训练条件下，FiberTune在跨越两个基准和两种架构（pi_0.5和OpenVLA-OFT）的六个受控仿真设置以及物理SO-101拾取放置任务中，均优于仅任务损失的微调；代表性提升包括长时域CALVIN ABC-to-D上SR(5)提高10.7个百分点，物理SO-101任务成功率从72.7%提升至78.1%。残差诊断显示，这些增益与探针过滤后的残差教师对齐度和有效秩增加一致，符合动作纤维动机。
英文摘要：
Action-supervised fine-tuning of vision-language-action (VLA) policies fits demonstrations effectively but constrains only the directions that change predicted actions, leaving visual structure consistent across action-equivalent states free to collapse. We formalize this as residual visual collapse along local action fibers and propose FiberTune, a training-time objective that preserves teacher-structured visual residuals without adding inference-time overhead. FiberTune uses an online action probe to estimate action-predictive feature directions, filters them from intermediate visual-token representations, and aligns the resulting probe-filtered residuals to a frozen visual teacher while regularizing their effective rank. Under identical training conditions, FiberTune improves over task-loss-only fine-tuning in every one of six controlled simulation settings spanning two benchmarks and two architectures (pi_0.5 and OpenVLA-OFT), as well as on physical SO-101 pick-place; representative gains include +10.7 percentage points SR(5) on long-horizon CALVIN ABC-to-D and physical SO-101 task success rising from 72.7% to 78.1%. Residual diagnostics show that these gains coincide with increased probe-filtered residual teacher alignment and effective rank, consistent with the action-fiber motivation.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
