trends.barnas.net

method · first seen 9 Oct 2026

zero-shot reinforcement learning

Also: ZSRL

Learning policies without task-specific training by inferring the policy at test time from reward-free pre-training.

Mentions per week

0119 May11 Aug10 NovPapersCodeDiscussion

PapersCodeDiscussion

Earliest sources

Latest sources

Mentioned together with

joint-embedding predictive architecturesparse representation learningbehavioral foundation modelmeta-reinforcement learning