AI Paper Atlas

Research map / Reinforcement learning / Offline RL

Continuity-Regularized Flow Matching for Offline Reinforcement Learning

Xiaocong Chen, Siyu Wang, Lina Yao

ICML 2026 · Reinforcement learning › Offline RL · approach: Behavior regularization for offline RL

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
add Jacobian (PDE) regularizer together with Beta timestep sampling (ablation)PQL · D4RL Hopper medium-expertnormalized score: 98.7 -> 111.8Improved
add Beta timestep sampling to the PDE-regularized model (vs uniform sampling) (ablation)PQL · D4RL HalfCheetah medium-expertnormalized score: 76.2 -> 90.2Improved
add PDE regularizer to Beta-sampled model (ablation)PQL · D4RL HalfCheetah medium-expertnormalized score: 84.3 -> 90.2Improved
add PDE regularizer with uniform time sampling (ablation)FQL-style flow policy · D4RL Gym-MuJoConormalized scoreHurt
add Beta timestep sampling without regularizer (ablation)PQL-PDE · D4RL Gym-MuJoConormalized scoreImproved
penalize full Jacobian Frobenius norm instead of divergence only (ablation)PQL · D4RL Gym-MuJoCo, harder Medium-Expert and Medium-Replaynormalized scoreImproved
penalize full Jacobian instead of divergence on Medium datasets (ablation)PQL · D4RL Gym-MuJoCo Mediumnormalized scoreNo clear effect
vary Beta shape alpha in [2, 5] and lambda in [0.01, 0.02] (ablation)PQL · D4RL Gym-MuJoConormalized scoreNo clear effect
PQL vs FQL on D4RL Gym-MuJoCo medium-expert HopperFQL · D4RL Hopper medium-expertnormalized score: 21 -> 112Improved
PQL on OGBench long-horizon navigation and manipulation vs FQLFQL · OGBench puzzle-4x4-playsuccess rate (%): 11 -> 16Improved
PQL on HalfCheetah medium vs CACPQL · D4RL HalfCheetah mediumnormalized score: 54 vs 69 (CAC)No clear effect

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