{"ID":6267119,"CreatedAt":"2026-07-10T01:11:38.759438437Z","UpdatedAt":"2026-07-13T01:02:08.706470581Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.08283","arxiv_id":"2607.08283","title":"TFP: Temporally Conditioned Memory-Fusion Policies for Visuomotor Learning","abstract":"Vision--Language--Action (VLA) policies such as $π_{0.5}$ and OpenVLA perform well on many manipulation tasks, but they are often reactive: the next action is predicted from the current observation, instruction, and proprioceptive state. This assumption breaks down in stage-dependent manipulation, where visually similar states may require different actions depending on latent task progress and previous interaction outcomes. We argue that such tasks require not only memory, but dynamics-aware belief updates: the policy should preserve task progress during stable or occluded phases and revise its belief near contact, release, or subgoal transitions. We introduce Temporally Conditioned Memory-Fusion Policies (TFP), a lightweight memory-action framework for VLA backbones. TFP maintains an episode-local task-progress belief with Liquid Time-Constant dynamics and injects the updated belief directly into the flow-matching action decoder through adaptive modulation. This lets temporally accumulated context shape the generated action chunk, rather than serving only as passive history context. With a 3.3B-parameter model, TFP improves the average success rate from \\(96.9\\%\\) to \\(98.75\\%\\) on LIBERO and from \\(91.4\\%\\) to \\(93.77\\%\\) on LIBERO-plus. On the memory-focused MIKASA ShellGameTouch diagnostic, TFP achieves success up to \\(75.0\\%\\). Mechanistic analyses show that write-gain changes near manipulation events are about \\(6\\times\\) larger than far non-event phases, and hidden-state interventions show that the belief causally modulates generated action chunks. These results suggest that compact, event-sensitive memory dynamics can improve VLA policies under occlusion, visual perturbation, and stage-dependent task structure.","short_abstract":"Vision--Language--Action (VLA) policies such as $π_{0.5}$ and OpenVLA perform well on many manipulation tasks, but they are often reactive: the next action is predicted from the current observation, instruction, and proprioceptive state. This assumption breaks down in stage-dependent manipulation, where visually simila...","url_abs":"https://arxiv.org/abs/2607.08283","url_pdf":"https://arxiv.org/pdf/2607.08283v1","authors":"[\"Yushen Liang\",\"Yue Peng\",\"Baosheng Jin\",\"Tianluo Zhang\",\"Xinyu Zhang\",\"Shuyi Zhou\",\"Zhuoran Chen\",\"Xinqi Liu\",\"Shenji Wan\"]","published":"2026-07-09T09:24:30Z","proceeding":"cs.RO","tasks":"[\"cs.RO\"]","methods":"[]","has_code":false}