Think Twice: Enhancing LLM Reasoning by Scaling Multi-round Test-time Thinking

• URL: http://arxiv.org/abs/2503.19855v1
• Date: Tue, 25 Mar 2025 17:19:38 GMT
• Title: Think Twice: Enhancing LLM Reasoning by Scaling Multi-round Test-time Thinking
• Authors: Xiaoyu Tian, Sitong Zhao, Haotian Wang, Shuaiting Chen, Yunjie Ji, Yiping Peng, Han Zhao, Xiangang Li,
• Abstract summary: We propose a simple yet effective test-time scaling approach Multi-round Thinking.<n>This method iteratively refines model reasoning by leveraging previous answers as prompts for subsequent rounds.<n>Experiments across multiple models, including QwQ-32B and DeepSeek-R1, consistently show performance improvements.
• Score: 16.441081996257576
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent advances in large language models (LLMs), such as OpenAI-o1 and DeepSeek-R1, have demonstrated the effectiveness of test-time scaling, where extended reasoning processes substantially enhance model performance. Despite this, current models are constrained by limitations in handling long texts and reinforcement learning (RL) training efficiency. To address these issues, we propose a simple yet effective test-time scaling approach Multi-round Thinking. This method iteratively refines model reasoning by leveraging previous answers as prompts for subsequent rounds. Extensive experiments across multiple models, including QwQ-32B and DeepSeek-R1, consistently show performance improvements on various benchmarks such as AIME 2024, MATH-500, GPQA-diamond, and LiveCodeBench. For instance, the accuracy of QwQ-32B improved from 80.3% (Round 1) to 82.1% (Round 2) on the AIME 2024 dataset, while DeepSeek-R1 showed a similar increase from 79.7% to 82.0%. These results confirm that Multi-round Thinking is a broadly applicable, straightforward approach to achieving stable enhancements in model performance, underscoring its potential for future developments in test-time scaling techniques. The key prompt: {Original question prompt} The assistant's previous answer is: <answer> {last round answer} </answer>, and please re-answer.

Related papers

• Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learning [231.11339402237903]
  We introduce Seed1.5-Thinking, capable of reasoning through thinking before responding. Seed1.5-Thinking achieves 86.7 on AIME 2024, 55.0 on Codeforces and 77.3 on GPQA. It demonstrates excellent reasoning abilities in STEM and coding.
  arXiv  Detail & Related papers  (2025-04-10T17:10:51Z)
• START: Self-taught Reasoner with Tools [51.38785489790888]
  We introduce START (Self-Taught Reasoner with Tools), a tool-integrated long Chain-of-thought (CoT) reasoning LLM.<n> START is capable of performing complex computations, self-checking, exploring diverse methods, and self-ging.<n>It significantly outperforms the base QwQ-32B and achieves performance comparable to the state-of-the-art open-weight model R1-Distill-Qwen-32B.
  arXiv  Detail & Related papers  (2025-03-06T17:11:51Z)
• An Empirical Study on Eliciting and Improving R1-like Reasoning Models [90.52239241349504]
  scaling RL training has become a central technique for implementing such reasoning models.<n>We demonstrate that our RL training approach consistently improves the Qwen2.5-32B base models.<n>We also explore the use of tool manipulation, finding that it significantly boosts the reasoning performance of large reasoning models.
  arXiv  Detail & Related papers  (2025-03-06T15:34:27Z)
• Dedicated Feedback and Edit Models Empower Inference-Time Scaling for Open-Ended General-Domain Tasks [7.686622572497795]
  Inference-Time Scaling has been critical to the success of recent models such as OpenAI o1 and DeepSeek R1.<n>We take inspiration from how humans make first attempts, ask for detailed feedback from others and make improvements based on such feedback.<n>We show that performance on Arena Hard, a benchmark of Arena Elo can be boosted by scaling the number of initial response drafts, effective feedback and edited responses.
  arXiv  Detail & Related papers  (2025-03-06T12:30:24Z)
• S*: Test Time Scaling for Code Generation [55.11863577956177]
  We propose S*, the first hybrid test-time scaling framework for code generation.<n>S* substantially improves the coverage and selection accuracy of generated code.
  arXiv  Detail & Related papers  (2025-02-20T09:18:53Z)
• Revisiting the Test-Time Scaling of o1-like Models: Do they Truly Possess Test-Time Scaling Capabilities? [61.85289698610747]
  We study whether o1-like large language models (LLMs) truly possess test-time scaling capabilities.<n>We find that longer CoTs of these o1-like models do not consistently enhance accuracy.<n>We propose Shortest Majority Vote, a method that combines parallel scaling strategies with CoT length characteristics.
  arXiv  Detail & Related papers  (2025-02-17T07:21:11Z)
• When Liebig's Barrel Meets Facial Landmark Detection: A Practical Model [87.25037167380522]
  We propose a model that is accurate, robust, efficient, generalizable, and end-to-end trainable. In order to achieve a better accuracy, we propose two lightweight modules. DQInit dynamically initializes the queries of decoder from the inputs, enabling the model to achieve as good accuracy as the ones with multiple decoder layers. QAMem is designed to enhance the discriminative ability of queries on low-resolution feature maps by assigning separate memory values to each query rather than a shared one.
  arXiv  Detail & Related papers  (2021-05-27T13:51:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.