RL training compute for logical reasoning follows a power law in proof depth whose exponent rises with logic expressiveness, and more expressive training yields larger gains on downstream benchmarks.
Reasoninggym: Reasoningenvironmentsforreinforcementlearningwithverifiable rewards
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5verdicts
UNVERDICTED 5representative citing papers
Fast-Slow Training combines slow parameter updates with fast context optimization to achieve up to 3x better sample efficiency, higher performance, less forgetting, and preserved plasticity in continual LLM learning.
AIPO trains LLMs to expand their reasoning capability boundary via active multi-agent interaction with Verify, Knowledge, and Reasoning agents during RLVR, using importance sampling and clipping to handle feedback, then drops the agents at inference.
QuickScope uses modified COUP Bayesian optimization to find truly difficult questions in dynamic LLM benchmarks more sample-efficiently than baselines while cutting false positives.
Systematic false positives in verifiers can cause RLVR training to reach suboptimal plateaus or collapse, with outcomes driven by error patterns rather than overall error rate.
citing papers explorer
-
Can RL Teach Long-Horizon Reasoning to LLMs? Expressiveness Is Key
RL training compute for logical reasoning follows a power law in proof depth whose exponent rises with logic expressiveness, and more expressive training yields larger gains on downstream benchmarks.
-
Learning, Fast and Slow: Towards LLMs That Adapt Continually
Fast-Slow Training combines slow parameter updates with fast context optimization to achieve up to 3x better sample efficiency, higher performance, less forgetting, and preserved plasticity in continual LLM learning.
-
AIPO: : Learning to Reason from Active Interaction
AIPO trains LLMs to expand their reasoning capability boundary via active multi-agent interaction with Verify, Knowledge, and Reasoning agents during RLVR, using importance sampling and clipping to handle feedback, then drops the agents at inference.
-
QuickScope: Certifying Hard Questions in Dynamic LLM Benchmarks
QuickScope uses modified COUP Bayesian optimization to find truly difficult questions in dynamic LLM benchmarks more sample-efficiently than baselines while cutting false positives.
-
Delay, Plateau, or Collapse: Evaluating the Impact of Systematic Verification Error on RLVR
Systematic false positives in verifiers can cause RLVR training to reach suboptimal plateaus or collapse, with outcomes driven by error patterns rather than overall error rate.