The paper introduces SpaRLax, a sparse relaxation method for reinforcement learning that enhances soft Chain-of-Thought reasoning in large language models. By treating soft tokens as differentiable intermediate steps and applying an adaptive entmax sparsity technique, SpaRLax stabilizes training and improves performance across nine reasoning benchmarks. The method retains multiple plausible candidates when uncertain while truncating noise at confident steps.
By Jiabin Fan, Matthias Lindemann, Zita Marinho, Andr\'e FT Martins, Lili Mou
arXiv:2509. 04027v4 Announce Type: replace Abstract: Test-time scaling, primarily manifested through multi-step Chain-of-Thought (CoT) reasoning via Reinforcement Learning (RL), has emerged as a pivotal paradigm for enhancing the reasoning capabilities of Large Language Models (LLMs).
By Zeyu Gan, Hao Yi, Yong Liu
arXiv:2509. 04027v3 Announce Type: replace Abstract: Test-time scaling, primarily manifested through multi-step Chain-of-Thought (CoT) reasoning via Reinforcement Learning (RL), has emerged as a pivotal paradigm for enhancing the reasoning capabilities of Large Language Models (LLMs).
By Zeyu Gan, Hao Yi, Yong Liu
CoTS: Competitive Token Selection for Latent Reasoning in Large Language Models proposes a training‑free framework that replaces binary latent reasoning decisions with continuous token‑level allocation. By using the probability competition between the two most likely tokens, CoTS dynamically balances discrete commitment and soft exploration at each reasoning step, eliminating the need for manually tuned thresholds or additional forward passes. Experiments on six reasoning benchmarks show that CoTS consistently outperforms standard Chain‑of‑Thought and other state‑of‑the‑art training‑free latent reasoning methods.
By Jie Yu, Yuxi Sun, Zhaoxu Li, Chunxiao Liu, Xudong Jiang
The paper reports that in on‑policy distillation for large language models, reasoning performance can be improved by supervising only a tiny fraction of generated tokens—sometimes just one or two tokens per reasoning trajectory, about 0.05% of all tokens. This sparse supervision consistently matches or exceeds full‑token training across nine teacher‑student setups on mathematical reasoning, and is also validated on coding reasoning, Llama models, and PPO‑based reinforcement learning with verifiable reward. The findings suggest that effective post‑training does not require token‑intensive supervision and may align more closely with natural learning processes that focus on critical reasoning steps.
By Zhishuai Liu, Xingzi Xu, Mehmet Saygin Seyfioglu, Pan Xu, Karim Bouyarmane
The paper introduces a method to enhance large language model (LLM) exploration in Reinforcement Learning with Verifiable Rewards (RLVR) by guiding the target model with partial reasoning trajectories from smaller, weaker language models. This weak-model guidance disrupts over‑confidence, preserves generative diversity, and mitigates entropy collapse without extra fine‑tuning or complex reward designs. Experiments on mathematical benchmarks show consistent improvements over vanilla RLVR, especially as the number of allowed attempts ($k$) increases, indicating broader reasoning coverage.
By Xingyu Shen, Huishuai Zhang, Peng Li, Yinchun Wang, Dongyan Zhao
arXiv:2606. 13862v1 Announce Type: cross Abstract: Long Chain-of-Thought (CoT) reasoning improves LLM problem-solving but is computationally expensive due to sequential token generation.
By Zheyang Xiong, Shivam Garg, Max Yu, Vaishnavi Shrivastava, Haoyu Zhao, Anastasios Kyrillidis, Dimitris Papailiopoulos
arXiv:2505. 12992v4 Announce Type: replace-cross Abstract: Inference-time scaling techniques have significantly bolstered the reasoning capabilities of large language models (LLMs) by harnessing additional computational effort at inference without retraining.
By Baohao Liao, Hanze Dong, Yuhui Xu, Doyen Sahoo, Christof Monz, Junnan Li, Caiming Xiong
arXiv:2602. 08324v5 Announce Type: replace Abstract: Chain-of-Thought (CoT) reasoning successfully enhances the reasoning capabilities of Large Language Models (LLMs), yet it incurs substantial computational overhead for inference.
By Yuntian Tang, Bohan Jia, Wenxuan Huang, Lianyue Zhang, Jiao Xie, Wenxi Li, Wei Li, Jie Hu, Xinghao Chen Rongrong Ji, Shaohui Lin
arXiv:2510. 08647v2 Announce Type: replace-cross Abstract: Recent developments have enabled advanced reasoning in Large Language Models (LLMs) via long Chain-of-Thought (CoT), trading efficiency during inference for performance.
By Chengzhengxu Li, Xiaoming Liu, Zhaohan Zhang, Shengchao Liu, Guoxin Ma, Yu Lan, Cong Wang, Chao Shen
The paper introduces Multi-Level Reasoning (MLR), a two-level stochastic framework that separates long chain-of-thought generation into a high-level planner producing structured step descriptors and a low-level executor generating detailed reasoning. MLR replaces single-policy reinforcement learning with an iterative Step‑DPO preference optimization that uses Twisted Sequential Monte Carlo for scalable, stepwise supervision, improving training stability and accuracy. Experiments on math, science, and logic benchmarks show that MLR outperforms both supervised fine‑tuning and strong RL baselines while maintaining robustness over extended reasoning horizons.
By Siheng Xiong, Ali Payani, Faramarz Fekri
arXiv:2608.30632v1 Announce Type: cross
Abstract: Reinforcement learning (RL), particularly RL with Verifiable Rewards (RLVR), has recently emerged as a central paradigm for enhancing large language...
By Outongyi Lv, Yuanwei Zhang, Xiaoqun Zhang