GCPO uses team-level credit assignment via determinant volume over reward-weighted semantic embeddings to promote non-redundant correct reasoning paths, improving both accuracy and diversity in LLM training.
Advancing language model reasoning through reinforcement learning and inference scaling.arXiv preprint arXiv:2501.11651
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
abstract
Large language models (LLMs) have demonstrated remarkable capabilities in complex reasoning tasks. However, existing approaches mainly rely on imitation learning and struggle to achieve effective test-time scaling. While reinforcement learning (RL) holds promise for enabling self-exploration, recent attempts yield modest improvements in complex reasoning. In this paper, we present T1 to scale RL by encouraging exploration and understand inference scaling. We first initialize the LLM using synthesized chain-of-thought data that integrates trial-and-error and self-verification. To scale RL training, we promote increased sampling diversity through oversampling. We demonstrate that T1 with open LLMs as its base exhibits inference scaling behavior and achieves superior performance on challenging math reasoning benchmarks. More importantly, we present a simple strategy to examine inference scaling, where increased inference budgets directly lead to T1's better performance without any additional verification.
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
Cross-modal agreement between chain-of-thought and program-of-thought reasoning enables self-consistency with only two LLM samples, reducing sampling cost by 9.3x while improving accuracy.
MARGO mitigates thinking-induced hallucination in large reasoning models by using mixed-mode GRPO rollout groups that compare thinking trajectories against same-model non-thinking references.
SAGE reshapes the reverse-KL anchor via guide function q(x,y) for controllable empirical support expansion, yielding gains in both pass@1 and pass@k on math reasoning benchmarks.
SCOPE-RL adds a regularization term built from high-temperature positive samples to quantitatively control entropy dynamics and maintain exploration in RL post-training of reasoning LLMs.
Self-Forcing++ scales autoregressive video diffusion to over 4 minutes by using self-generated segments for guidance, reducing error accumulation and outperforming baselines in fidelity and consistency.
UEC-RL improves RL reasoning performance in LLMs and VLMs by activating exploration on hard prompts and stabilizing entropy, delivering a 37.9% relative gain over GRPO on Geometry3K.
EEPO uses sample-then-forget rollouts with adaptive unlearning to boost exploration in RLVR, delivering relative gains of 24.3% on Qwen2.5-3B, 33.0% on Llama3.2-3B-Instruct, and 10.4% on Qwen3-8B-Base over GRPO across five reasoning benchmarks.
The paper unifies perspectives on Long CoT in reasoning LLMs by introducing a taxonomy, detailing characteristics of deep reasoning and reflection, and discussing emergence phenomena and future directions.
citing papers explorer
-
Breaking $\textit{Winner-Takes-All}$: Cooperative Policy Optimization Improves Diverse LLM Reasoning
GCPO uses team-level credit assignment via determinant volume over reward-weighted semantic embeddings to promote non-redundant correct reasoning paths, improving both accuracy and diversity in LLM training.
-
Self-Consistency from Only Two Samples: CoT-PoT Ensembling for Efficient LLM Reasoning
Cross-modal agreement between chain-of-thought and program-of-thought reasoning enables self-consistency with only two LLM samples, reducing sampling cost by 9.3x while improving accuracy.
-
Mitigating Factual Hallucination in Large Reasoning Models via Mixed-Mode Advantage Regularization
MARGO mitigates thinking-induced hallucination in large reasoning models by using mixed-mode GRPO rollout groups that compare thinking trajectories against same-model non-thinking references.
-
SAGE: Shaping Anchors for Guided Exploration in RLVR of LLMs
SAGE reshapes the reverse-KL anchor via guide function q(x,y) for controllable empirical support expansion, yielding gains in both pass@1 and pass@k on math reasoning benchmarks.
-
SCOPE-RL: Stable and Quantitative Control of Policy Entropy in RL Post-Training
SCOPE-RL adds a regularization term built from high-temperature positive samples to quantitatively control entropy dynamics and maintain exploration in RL post-training of reasoning LLMs.
-
Self-Forcing++: Towards Minute-Scale High-Quality Video Generation
Self-Forcing++ scales autoregressive video diffusion to over 4 minutes by using self-generated segments for guidance, reducing error accumulation and outperforming baselines in fidelity and consistency.
-
Targeted Exploration via Unified Entropy Control for Reinforcement Learning
UEC-RL improves RL reasoning performance in LLMs and VLMs by activating exploration on hard prompts and stabilizing entropy, delivering a 37.9% relative gain over GRPO on Geometry3K.
-
EEPO: Exploration-Enhanced Policy Optimization via Sample-Then-Forget
EEPO uses sample-then-forget rollouts with adaptive unlearning to boost exploration in RLVR, delivering relative gains of 24.3% on Qwen2.5-3B, 33.0% on Llama3.2-3B-Instruct, and 10.4% on Qwen3-8B-Base over GRPO across five reasoning benchmarks.
-
Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models
The paper unifies perspectives on Long CoT in reasoning LLMs by introducing a taxonomy, detailing characteristics of deep reasoning and reflection, and discussing emergence phenomena and future directions.