AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Hindsight Relabeling is All You Need for Reach-Avoid Learning

Kevin Li, Marinka Zitnik

NeurIPS 2026 · Reinforcement learning › Visual and representation RL · approach: In-context and adaptive RL

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes a hindsight relabeling approach for offline goal-conditioned RL on reach-avoid tasks, aiming to handle avoid regions specified flexibly at evaluation time.

Problem

Offline goal-conditioned RL for reach-avoid tasks, where existing methods encode avoid-region information into an augmented state space and cost function, preventing flexible specification of novel avoid regions at evaluation time.

Approach

Uses hindsight relabeling to handle reach-avoid objectives, so avoid-region information need not be baked into the augmented state or cost function; the abstract's full method description is truncated.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
encode goal and avoid-regions as prompt tokens (RADT) instead of augmented state with cost function, evaluated zero-shot on larger avoid boxesRADT vs retrained AM-Lag · FetchReachObstacle, box width 0.24MNC: 0.457 (AM-Lag retrained) -> 0.409 (RADT zero-shot)Improved
use RADT zero-shot on more avoid-regions than seen in trainingRADT · UMazeObstacle, 5 avoid-regionsMNC (1e-2): 10.14 (AM-Lag retrained) -> 8.75Improved
RADT vs AM-Lag in-distribution on FetchReachObstacleRADT · FetchReachObstacle, box width 0.16MNC: 0.011 (AM-Lag) vs 0.049 (RADT)Hurt
RADT zero-shot with more avoid-regions in AMaze/HMaze where AM-Lag is in-distributionRADT · HMazeObstacle, 7 avoid-regionsMNC (1e-2): 6.5 vs 6.03 (AM-Lag)No clear effect
add avoid success indicator, AS, GS tokens, relabeling and attention boosting to prompt (ablation)RADT · ablation suite (FetchReachObstacle / Maze)MNC and SRImproved
mix expert-policy trajectories into random training data (ablation)RADT · OOD box sizesMNC and SRMixed
mix expert-policy trajectories into random training data (ablation)RADT · in-distribution box sizesperformanceNo clear effect
use spherical (centroid+radius) avoid-region representation instead of box coordinates (ablation)RADT · FetchReachObstaclereach-avoid performanceNo clear effect
zero-shot RADT up to 20 avoid-regions vs retrained AM-LagRADT · MazeObstacle, up to 20 avoid-regionsSRImproved

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