← Back to all news
arXiv Machine Learning By Amit Thakur, Mukesh Singhal

Permutation Robustness Is Not Enough: Action Collapse in Multi-Agent Transformer Policies

Read the original on arXiv Machine Learning →

The paper examines how transformer policies, which process agents as ordered token sequences, perform in multi‑agent robot learning where agent teams are unordered. It finds that low permutation error can mask action collapse, where all agents choose the same action, and proposes additional diagnostics such as action diversity and same‑action fraction. Experiments show that while a PPO‑ID baseline avoids collapse, it remains order‑sensitive, and that strong equivariance regularization can still cause homogeneous behavior; a weak penalty improves robustness and preserves diversity for three‑agent teams, but four‑agent teams need much smaller regularization weights.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Jul 30

Interpretable GOHR Agents via Sparse Autoencoders

arXiv:2607. 25132v2 Announce Type: replace Abstract: A central challenge in interpreting learned decision-making systems is to determine whether their internal representations contain concepts that help explain their behavior.

By Shiwei Tan, Yusong Zhao, Weiyi Qin, Wentian Wang, Jacob Feldman, Lazaros K. Gallos, Paul B. Kantor, Vladimir Menkov, Hao Wang
llmsragagentssafety
More like this →
arXiv Machine Learning
Sep 10

Distributed Dexterous Manipulation with Spatially Conditioned Multi-Agent Transformers

The paper introduces Distributed Dexterous Manipulation (DDM), a challenging control problem involving 64 soft delta robots arranged in an 8x8 grid. It presents a framework using spatially conditioned Multi-Agent Transformers (MATs) with adaptive layer norm, spatial contrastive embeddings, and a behavior cloning method fine‑tuned by Soft Actor Critic. Experiments demonstrate that MATs refine actions through stacked attention blocks, enabling long‑horizon planar manipulation in simulation and real‑world settings, while an action‑selection strategy reduces robot usage by about 65% and lowers wear‑and‑tear, achieving an average error of ~1.5 cm.

By Sarvesh Patil
llmsragagentsroboticsfine-tuning
More like this →
arXiv AI
6d ago

Cooperative Multi-Agent Vision-Language-Action Models via Reinforced Fine Tuning

arXiv:2609.36588v1 Announce Type: cross Abstract: We study reinforcement learning (RL) methods for cooperative multi-agent Vision-Language-Action (VLA) models. This problem is challenging because VLA...

By Ruixiao Xu, Wong Lik Hang Kenny, Zhiqian Liu, Jianing Guo, Hanxiao Li, Kejian Shi, Shuning Zhang, Pu Feng, Yongjia Ma, Yuqing Ma, Kai Chen, Qi Dou, Yaodong Yang, Xianglong Liu, Simin Li
agentsreinforcement-learningroboticsfine-tuningmultimodal
More like this →
Hugging Face Trending Papers
Jul 30

RoboBRIDGE: A Modular Framework for Bridging Policies to Robust Real-World Robotic Agents

Vision-Language-Action (VLA) models have attracted growing interest as a scalable approach to robotic manipulation. While these models are effective action predictors, deploying them as robotic agents exposes critical gaps: no mechanism for failure recovery, inconsistent execution over long horizons, and limited robustness to shifts in observations, tasks, or embodiments.

agentsroboticsfine-tuningmultimodal
More like this →