PaceReasoner is a label‑free, test‑time reinforcement learning framework that improves short‑mode accuracy by leveraging a model’s long responses. It shares a trainable policy between long and short modes, using progressive, confidence‑weighted supervision and a behavior reward that shifts from encouraging short exploration to rewarding agreement with long answers. Across three backbones and four benchmarks, it raises short‑mode accuracy by 7.4–13.5 percentage points while keeping responses under 25 tokens on average.
arXiv:2606. 18831v1 Announce Type: cross Abstract: Long-context reasoning is an essential capability for large language models, particularly when they are deployed as autonomous agents that must reason over lengthy trajectories.
By Xiaoyue Xu, Sikui Zhang, Xiaorong Wang, Xu Han, Chaojun Xiao
arXiv:2507.21931v2 Announce Type: replace-cross
Abstract: Large Language Models (LLMs) often produce plausible but poorly-calibrated answers, limiting their reliability on reasoning-intensive tasks....
By Carel van Niekerk, Renato Vukovic, Benjamin Ruppik, Hsien-chin Lin, Shutong Feng, Milica Ga\v{s}i\'c
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:2609.37119v1 Announce Type: cross
Abstract: Recent approaches to reinforcement learning (RL) post-training for large language models increasingly remove the critic to reduce training instabilit...
By Hongyang Li, Xiao Li, Caesar Wu, Said Mammar, Gr\'egoire Danoy, Pascal Bouvry
arXiv:2511. 02130v2 Announce Type: replace-cross Abstract: We propose Re-FORC, an adaptive reward prediction method that, given a query, enables prediction of the expected future rewards as a function of the number of future thinking tokens.
By Renos Zabounidis, Aditya Golatkar, Michael Kleinman, Alessandro Achille, Wei Xia, Stefano Soatto
arXiv:2606. 11209v1 Announce Type: cross Abstract: Visual question answering increasingly requires multi-step reasoning.
By Jingpei Wu, Xiao Han, Weixiang Shen, Boer Zhang, Zifeng Ding, Volker Tresp
arXiv:2602. 06960v3 Announce Type: replace-cross Abstract: Large reasoning models achieve strong performance by scaling inference-time chain-of-thought, but this paradigm suffers from quadratic cost, context length limits, and degraded reasoning due to lost-in-the-middle effects.
By Yuchen Yan, Liang Jiang, Jin Jiang, Shuaicheng Li, Zujie Wen, Zhiqiang Zhang, Jun Zhou, Jian Shao, Yueting Zhuang, Yongliang Shen
The paper introduces Thinking-Free Policy Initialization (TFPI), a simple adaptation to Reinforcement Learning with Verifiable Reward (RLVR) that removes explicit thinking content during training and inference. TFPI bridges long Chain-of-Thought distillation with standard RLVR, reducing token usage and improving performance. Experiments show that TFPI accelerates RL convergence, raises performance ceilings, and yields token‑efficient reasoning models, enabling a 4B model to achieve 89.0% accuracy on AIME24 and 65.5% on LiveCodeBench with under 4K H20 hours.
By Xin Xu, Clive Bai, Kai Yang, Tianhao Chen, Yang Wang, Saiyong Yang, Can Yang
arXiv:2608. 03119v1 Announce Type: new Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) improves LLM reasoning but typically relies on ground-truth (GT) answers, limiting scalability.
By Yongshi Ye, Liang Zhang, Yidong Chen, Xiaodong Shi, Biao Fu
arXiv:2604. 22565v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) can reason well, yet often miss decisive evidence when it is buried in long, noisy contexts.
By Shaoang Li, Yanhang Shi, Yufei Li, Mingfu Liang, Xiaohan Wei, Yunchen Pu, Fei Tian, Chonglin Sun, Frank Shyu, Luke Simon, Sandeep Pandey, Xi Liu, Jian Li
Rationale-Guided Policy Optimization (RGPO) is a reinforcement‑learning framework that adaptively uses ground‑truth rationale information to scaffold a language model’s reasoning process. Instead of treating reference solutions as fixed imitation targets, RGPO temporarily incorporates rationales to help the model generate better responses, then reverts to unguided learning with higher‑reward, model‑generated solutions. Experiments in both language‑only and vision‑language tasks show that RGPO consistently outperforms RLVR baselines, with ablation studies confirming that adaptive rationale guidance is a key factor in its success.
By Hoang Phan, Minh Pham, Chau Pham, Chinmay Hegde, Trung Le, Qi Lei