CERO uses Beta posteriors and Fenchel-dual online optimization to adaptively allocate a fixed rollout budget across prompts and epochs in LLM RL, outperforming fixed-allocation GRPO on math reasoning benchmarks.
hub
Sample more to think less: Group filtered policy optimization for concise reasoning.arXiv preprint arXiv:2508.09726
15 Pith papers cite this work. Polarity classification is still indexing.
15
Pith papers citing it
hub tools
citation-role summary
background 3
baseline 1
citation-polarity summary
representative citing papers
LEAD uses online adaptive mechanisms including Potential-Scaled Instability and symmetric efficiency rewards based on correct rollouts to achieve higher accuracy-efficiency scores with substantially shorter reasoning outputs than base models on math benchmarks.
CoDistill-GRPO lets small and large models mutually improve via co-distillation in GRPO, raising small-model math accuracy by over 11 points while cutting large-model training time by about 18%.
Post-Reasoning boosts LLM accuracy by reversing the usual answer-after-reasoning order, delivering mean relative gains of 17.37% across 117 model-benchmark pairs with zero extra cost.
This survey introduces the Generate-Filter-Control-Replay (GFCR) taxonomy to structure rollout pipelines for RL-based post-training of reasoning LLMs.
Hidden-Align adds an auxiliary loss to align hidden states of correct reasoning paths at the pre-answer token in RLVR, improving pass@1 by 3.8-6.2 points over DAPO on eight math benchmarks for Qwen3 models of 1.7B-14B scale.
Pilot-Commit estimates per-prompt informativeness via a pilot stage and skips low-variance prompts, matching baseline accuracy with up to 4.0x fewer cumulative rollouts than DAPO on math reasoning tasks.
MEMENTO trains LLMs to segment reasoning into blocks, generate mementos as dense summaries, and reason forward using only mementos and KV states, cutting peak KV cache by ~2.5x while preserving benchmark accuracy.
RTT bridges response-level rubrics to token-level rewards via a relevance discriminator and intra-sample group normalization, yielding higher instruction and rubric accuracy than baselines.
GDPO decouples per-reward normalization in multi-reward RL to avoid advantage collapse and improve convergence over GRPO on tool-calling, math, and coding tasks.
Entropy After </Think> (EAT) enables early exiting in reasoning LLMs by tracking entropy stabilization after a </think> token, cutting token use 12-22% on MATH500 and AIME2025 with no accuracy loss.
TrOPD stabilizes on-policy distillation for LLMs with trust-region learning, outlier estimation, and off-policy guidance, outperforming prior OPD methods on reasoning and code benchmarks.
RLVR exhibits implicit reward overfitting to training data and optimizes heavy-tailed singular spectra with rank-1 focus on reasoning capability.
A survey organizing techniques to achieve efficient reasoning in LLMs by shortening chain-of-thought outputs.
A textbook-derived neuroscience knowledge graph supplies synthetic multi-hop QA supervision and RL rewards to fine-tune a small LM claimed to exceed larger general models on expert reasoning.
citing papers explorer
-
Rubrics to Tokens: Bridging Response-level Rubrics and Token-level Rewards in Instruction Following Tasks
RTT bridges response-level rubrics to token-level rewards via a relevance discriminator and intra-sample group normalization, yielding higher instruction and rubric accuracy than baselines.
-
Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models
A survey organizing techniques to achieve efficient reasoning in LLMs by shortening chain-of-thought outputs.