paper-with-me

홈 › Papers

Making Small Language Models Efficient Reasoners: Intervention, Supervision, Reinforcement

2025-05-12 · Xuechen Zhang, Zijian Huang, Chenshun Ni, Ziyang Xiong, Jiasi Chen, Samet Oymak

Recent research enhances language model reasoning by scaling test-time compute via longer chain-of-thought traces. This often improves accuracy but also introduces redundancy and high computational cost, especially for small language models distilled with supervised fine-tuning (SFT). In this work, we propose new algorithms to improve token-efficient reasoning with small-scale models by effectively trading off accuracy and computation. We first show that the post-SFT model fails to determine the optimal stopping point of the reasoning process, resulting in verbose and repetitive outputs. Verbosity also significantly varies across wrong vs correct responses. To address these issues, we propose two solutions: (1) Temperature scaling (TS) to control the stopping point for the thinking phase and thereby trace length, and (2) TLDR: a length-regularized reinforcement learning method based on GRPO that facilitates multi-level trace length control (e.g. short, medium, long reasoning). Experiments on four reasoning benchmarks, MATH500, AMC, AIME24 and OlympiadBench, demonstrate that TS is highly effective compared to s1's budget forcing approach and TLDR significantly improves token efficiency by about 50% with minimal to no accuracy loss over the SFT baseline. Moreover, TLDR also facilitates flexible control over the response length, offering a practical and effective solution for token-efficient reasoning in small models. Ultimately, our work reveals the importance of stopping time control, highlights shortcomings of pure SFT, and provides effective algorithmic recipes.

📄 PDF Abstract BibTeX arXiv:2505.07961

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

SFT Shrink and Fine-Tune, or SFT, is a type of distillation that avoids explicit distillation by copying parameters to a student student model and then fine-tuning.…
TS Spatio-temporal features extraction that measure the stabilty. The proposed method is based on a compression algorithm named Run Length Encoding. The workflow of the method is…

Similar Papers 제목 키워드 기반

SAPO: Self-Adaptive Process Optimization Makes Small Reasoners Stronger

2026-01-28 · Kaiyuan Chen, Guangmin Zheng, Jin Wang, Xiaobing Zhou 외 arxiv

Existing self-evolution methods overlook the influence of fine-grained reasoning steps, which leads to the reasoner-verifier gap. The computational inefficiency of Monte Carlo (MC) process supervision further exacerbates…

Clarification as Supervision: Reinforcement Learning for Vision-Language Interfaces

2025-09-30 · John Gkountouras, Ivan Titov arxiv

Recent text-only models demonstrate remarkable mathematical reasoning capabilities. Extending these to visual domains requires vision-language models to translate images into text descriptions. However, current models, t…

Reinforcement LearningMathematical Reasoning

Incentivizing Reasoning from Weak Supervision

2025-05-26 · Yige Yuan, Teng Xiao, Shuchang Tao, Xue Wang 외

Large language models (LLMs) have demonstrated impressive performance on reasoning-intensive tasks, but enhancing their reasoning abilities typically relies on either reinforcement learning (RL) with verifiable signals o…

reinforcement-learningReinforcement LearningReinforcement Learning (RL)

Explanations from Large Language Models Make Small Reasoners Better

2022-10-13 · Shiyang Li, Jianshu Chen, Yelong Shen, Zhiyu Chen 외

Integrating free-text explanations to in-context learning of large language models (LLM) is shown to elicit strong reasoning capabilities along with reasonable explanations. In this paper, we consider the problem of leve…

Explanation GenerationIn-Context LearningMulti-Task Learning

ProxyThinker: Test-Time Guidance through Small Visual Reasoners

2025-05-30 · Zilin Xiao, Jaywon Koo, Siru Ouyang, Jefferson Hernandez 외

Recent advancements in reinforcement learning with verifiable rewards have pushed the boundaries of the visual reasoning capabilities in large vision-language models (LVLMs). However, training LVLMs with reinforcement fi…

Visual Reasoning