Median Temporal Ensembling: Training-Free Robust Aggregation for Action-Chunked Visuomotor Policies

arXiv:2609.27167 · cs.RO, cs.LG · Submitted 2026-09-22 · Read on arXiv

cs.RO, cs.LG

Submitted: 2026-09-22

Updated: 2026-09-22

Comments: 9 pages, 2 figures, 6 tables. Project page: https://avalon-s.github.io/MedianTE/

Project page: https://avalon-s

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Aggregation — 10× in this paper · explained in 1 episode
• Robustness — 10× in this paper · explained in 5 episodes
• Fine-Tuning — 6× in this paper · explained in 5 episodes
• Action Chunking — 5× in this paper · explained in 1 episode
• Consistency — 5× in this paper · explained in 3 episodes
• Redundancy — 5× in this paper · explained in 2 episodes
• Adversarial Attacks — 3× in this paper · explained in 1 episode
• International Conference on Machine Learning — 2× in this paper
• Adversarial Examples — 1× in this paper · explained in 2 episodes
• Adversarial Robustness — 1× in this paper · explained in 3 episodes
• Adversarial Training — 1× in this paper · explained in 1 episode
• Benchmarking — 1× in this paper · explained in 3 episodes
• Conference on Robot Learning — 1× in this paper
• Failure Modes — 1× in this paper · explained in 3 episodes
• Federated Learning — 1× in this paper · explained in 14 episodes
• International Conference on Learning Representations — 1× in this paper
• Model-Agnostic — 1× in this paper · explained in 2 episodes
• Pairwise Comparisons — 1× in this paper · explained in 2 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Transformer — 1× in this paper · explained in 2 episodes

Sources

• Temporal Action Selection for Action Chunking
• Model-agnostic Adversarial Attack and Defense for Vision-Language-Action Models
• Why Does Action Chunking Improve Behavioral Cloning Performance in Robotic Control?
• VLA-FAIL: Efficient Task Failure Detection for Finetuned Vision-Language-Action Models
• Rewind-IL: Online Failure Detection and State Respawning for Imitation Learning
• Adversarial Attacks on Neural Network Policies
• Adversarial examples in the physical world
• On Evaluating Adversarial Robustness

Related papers

• FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
• MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
• RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
• ARC: A Reasoning Recipe for Robot Foundation Models
• HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
• APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies