AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

Value Flows

Perry Dong, Chongyi Zheng, Chelsea Finn, Dorsa Sadigh, Benjamin Eysenbach

ICLR 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Distributional value learning · 14 citations (OpenAlex)

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
model return distribution with flow matching instead of C51 bins or CODAC quantilesValue Flows · OGBench scene-play-singletask-task21-Wasserstein distance: 3x lower than best baselineImproved
add BCFM regularization to distributional flow matching objective, with tuned coefficient (ablation)Value Flows · OGBench puzzle-3x3success rate: 2.6x with best lambdaImproved
tune BCFM regularization coefficient to best value (ablation)Value Flows · OGBench puzzle-3x3sample efficiency: 2x sample efficiencyImproved
reweight flow-matching loss with return-variance confidence weight at appropriate temperature (ablation)Value Flows · two OGBench state taskssuccess rate: 60% higher on averageImproved
use Value Flows instead of best baseline on challenging state-based tasksValue Flows · OGBench harder state-based domainssuccess rate: 1.6x higher than best baselineImproved
use Value Flows on image-based tasksValue Flows · OGBench visual taskssuccess rate: 1.24x over best baselineImproved
use Value Flows for online fine-tuning on puzzle-4x4-playValue Flows · OGBench puzzle-4x4-play offline-to-onlinesuccess rate: 15% higher than prior offline-to-online algorithmsImproved
Value Flows versus scalar-Q methods on D4RL adroitValue Flows · D4RL adroit (12 tasks)success rate: 50 vs 59 (ReBRAC)No clear effect
use naive distributional conditional flow matching loss without bootstrapping (ablation)Value Flows · initial experiments on some tasksperformanceHurt

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