The paper investigates whether the high costs of training chain-of-thought reasoning models can be reduced through algorithmic design. It introduces an autocurriculum approach that lets the model select which problems to focus on during training, showing that this method provably improves both supervised fine‑tuning and reinforcement learning. For supervised fine‑tuning, autocurriculum requires exponentially fewer reasoning demonstrations by targeting prompts where the model struggles, while for reinforcement learning it decouples computational cost from the quality of the reference model, making the burn‑in cost nearly independent of target accuracy.
By Nived Rajaraman, Audrey Huang, Miro Dudik, Robert Schapire, Dylan J. Foster, Akshay Krishnamurthy
arXiv:2606. 03965v1 Announce Type: cross Abstract: Large language models improve final-answer accuracy through extended chain-of-thought reasoning, but often spend tokens inefficiently and offer little inference-time control.
By Yu Xia, Zhouhang Xie, Xin Xu, Byungkyu Kang, Prarit Lamba, Xiang Gao, Julian McAuley
The paper demonstrates that fine‑tuning reasoning models to predict their own confidence at intermediate steps—using only 600 self‑supervised examples—substantially improves inference efficiency. Without adding any explicit stopping or length penalties, the models generate up to 25 % fewer tokens while maintaining accuracy on mathematical, scientific, and coding benchmarks across several architectures. The study finds that confidence supervision preserves the models’ high‑level reasoning structure rather than merely suppressing specific behaviors.
By Parsa Hosseini, Akasha Tigalappanavara, Sumit Nawathe, Chenrui Fan, Sourya Basu, Genta Indra Winata, Anirban Das, Soheil Feizi, Nima Chitsazan
arXiv:2605. 03862v4 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards has become a common way to improve explicit reasoning in large language models, but final-answer correctness alone does not reveal whether the reasoning trace is faithful, reliable, or useful to the model that consumes it.
By Tianyang Han, Hengyu Shi, Junjie Hu, Xu Yang, Zhiling Wang, Junhao Su
arXiv:2607. 18100v1 Announce Type: new Abstract: Extended reasoning has become standard for frontier Large Language Models (LLMs), yet the trajectories these models produce remain largely uncontrollable.
By Sheldon Yu, Tong Yu, Xunyi Jiang, Rohan Surana, Gagan Mundada, Sungchul Kim, Lina Yao, Julian McAuley, Junda Wu
arXiv:2508. 04460v2 Announce Type: replace Abstract: Large Reasoning Models (LRMs) can exhibit step-by-step reasoning, reflection, and backtracking, but these behaviors are often unregulated, leading to overthinking.
By Rui Ha, Rui Pu, Chaozhuo Li, Li Sun, Sen Su
arXiv:2509.14093v3 Announce Type: replace-cross
Abstract: Chain-of-Thought (CoT) prompting can substantially improve the reasoning ability of large language models (LLMs), but it often comes with hig...
By Kerui Huang, Shuhan Liu, Xing Hu, Tongtong Xu, Lingfeng Bao, Xin Xia
arXiv:2604.02007v3 Announce Type: replace
Abstract: Building general-purpose reasoning models using reinforcement learning with verifiable rewards (RLVR) across diverse domains has been widely adopte...
By Rafael Pardinas, Ehsan Kamalloo, David Vazquez, Alexandre Drouin
The paper introduces a method to control reasoning behaviors in Mixture-of-Experts (MoE) language models by labeling sentence-level segments with six distinct reasoning behaviors (planning, execution, restatement, fact, reflection, and conclusion). Linear probes on MoE router logits can predict these behaviors nearly as accurately as probes on the full hidden state, enabling the authors to edit router scores during inference to reduce overthinking. This intervention cuts generated tokens by a median of 30.4% (up to 63.3%) across multiple models and benchmarks without significant loss in accuracy, and often improves it.
By Shengjie Wang, Mohsen Fayyaz, Nanyun Peng
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:2603.00578v2 Announce Type: replace-cross
Abstract: Long chain-of-thought~(CoT) reasoning improves the problem-solving capability of large reasoning models~(LRMs), but often incurs substantial...
By Jie Cao, Tianwei Lin, Juekai Lin, Zhenxuan Fan, Yunqi Cao, Bo Yuan, Ziyuan Zhao, Rolan Yan, Wenqiao Zhang, Siliang Tang
TRACES (Tagging Reasoning Steps for Adaptive Cost‑Efficient Early‑Stopping) is a lightweight framework that tags reasoning steps of large‑language models in real time, enabling adaptive, cost‑efficient early stopping during inference. By monitoring the types of steps generated, the method identifies when models shift their reasoning after arriving at a correct answer, allowing for interpretable stopping criteria. Experiments on mathematical reasoning benchmarks (MATH500, GSM8K, AIME) and knowledge benchmarks (MMLU, GPQA) show token reductions of 20–50% while preserving accuracy, with more conservative thresholds needed for harder tasks such as BeyondAIME and IMO AnswerBench.
By Yannis Belkhiter, Seshu Tirupathi, Giulio Zizzo, John D. Kelleher