Early-Signal Detector

method · first seen 12 Oct 2026

test-time reinforcement learning

Also: TTRL, test-time RL

Adapting a language model with reinforcement learning on unlabeled test problems using pseudo-labels from sampled responses.

Mentions per week

0319 May11 Aug10 NovPapersCodeDiscussion

PapersCodeDiscussion

Rank 137 on 11 Nov 2026 · 7 mentions in 28 days · 5 independent groups · papers · first seen 4 weeks ago · acceleration +1.66

Earliest sources

Latest sources

Mentioned together with

group relative policy optimizationreinforcement learning with verifiable rewardschain-of-thoughtmajority voting pseudo-labelingentropy-based rewardpolicy entropy collapsedirect preference optimizationmajority-vote reward