The paper introduces AnchorQ, a method that improves sample-efficient reinforcement learning for policy refinement by using flow-reversed action anchors. AnchorQ maps reference behaviors to state-conditioned latent anchors, samples local particles around these anchors, and employs value-guided sampling to explore promising refinements. Experiments on simulated and real-world manipulation tasks show that AnchorQ outperforms state-of-the-art post-training baselines, achieving a 37% higher real-world success rate and 97% greater simulated task throughput within the same interaction budget.
By Carl Qi, Weijun Dong, Xiaojie Wang, Peiyi Chen, Rahul Sajnani, Mingyuan Zhang, Amy Zhang, Jose Barreiros, Manikatan Nambi, Lillian J. Ratliff, Dylan Randle
arXiv:2607. 04265v1 Announce Type: cross Abstract: World-action (WA) models can generate long-horizon action chunks for general-purpose robotic manipulation, but they remain vulnerable to calibration, perception, and contact-dynamics errors in real-world precision tasks, often failing in the final few millimeters of alignment or insertion.
By Angen Ye, Weijie Ke, Xiaofeng Wang, Xinze Chen, Chaojun Ni, Guosheng Zhao, Boyuan Wang, Zheng Zhu, Junjie Xie, Dapeng Zhang
Video Prediction Policy 2 (VPP2) is a new world action model that improves zero‑shot generalization for both video prediction and action generation. It achieves this by pretraining a large, diverse manipulation video dataset with event‑level supervision, then distilling the model into a single‑step visual planner and adding a mixture‑of‑transformers action module. Experiments show VPP2 outperforms leading baselines on open‑ended video prediction, real‑world zero‑shot manipulation, and several challenging benchmarks.
By Yanjiang Guo, Haodong Yan, Zhide Zhong, Zhongru Zhang, Qingyuan Yang, Qingzhou Lu, Xiaoyu Chen, Yen-Jen Wang, Shuying Deng, Chenghan Yang, Puzhen Yuan, Chenxin Liu, Tun Ban, Xiang Zhu, Yichen Liu, Kun Feng, Haoang Li, Jianyu Chen
The paper introduces SPACE, a method for enabling large language model agents to emit variable-length action chunks in long-horizon tasks. By distilling chunk-boundary supervision from programmatic skills derived from successful trajectories, SPACE overcomes the tendency of agents to either act one step at a time or commit to overly long sequences. Experiments on ALFWorld and ScienceWorld demonstrate that SPACE raises success rates by 7.0%–31.3% and cuts LLM decision rounds by up to 78.9%.
By Yanting Yang, Can Jin, Jinman Zhao, Jiahao Wu, Yang Zhou, Zhepeng Wang, Zhendong Wang, Mu Zhou, Dimitris N. Metaxas
DriftOPD is a teacher‑free, rollout‑free framework that performs sequence‑level on‑policy distillation of continuous Vision‑Language‑Action (VLA) action experts. It decomposes the sequence‑level reverse‑KL divergence into a chunk‑level reverse‑KL term and a future‑potential term, optimizing them with a one‑step drifting objective and a Q‑function critic learned from offline demonstrations. Experiments on multiple VLA architectures in simulation and real‑world manipulation show that DriftOPD outperforms existing one‑step distillation baselines while matching the task success of multi‑step teacher policies.
By Youngjun Jun, Kyumin Choi, Youngmin Kim, Seonghyun Jin, Sunwoo Park, Jangho Park, Jong Chul Ye
Large language model (LLM) agents for long-horizon interactive tasks typically follow a ReAct-style protocol, issuing one primitive action per LLM round. While this enables frequent replanning, it is...