AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Demystifying Robot Diffusion Policies: Action Memorization and a Simple Lookup Table Alternative

Chengyang He, Xu Liu, Gadiel Mark Sznaier Camps, Joseph Bruno, Guillaume Adrien Sartoretti, Mac Schwager

ICLR 2026 · Reinforcement learning › Visual and representation RL · approach: Skill learning and diffusion policies

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use ALT instead of Diffusion PolicyDiffusion Policy · Cup grasping real robotinference time: ~2.65 s -> ~0.009 sImproved
Use ALT instead of Diffusion PolicyDiffusion Policy · Cup grasping OOD scenariosretrieval correctness: DP fails OOD4, OOD5; ALT w/o p gamma=0.75 fails to flag OOD7Mixed
Use contrastive latent search instead of pixel K-D treeK-D Tree · Cup graspinginference time: ~0.09 s -> ~0.009 sImproved
Add end-effector pose input p to ALT with gamma=0.9 (ablation)ALT · OOD scenariosOOD flag/retrievalImproved
Lower OOD threshold gamma from 0.9 to 0.75 (ablation)ALT w/o p · OOD scenariosretrieval correctnessMixed
Early stopping at best validation loss (ablation)Diffusion Policy · training set action MSEaction MSE: around 1800 remainsHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy