Teaching Thinking Models to Reason with Tools: A Full-Pipeline Recipe for Tool-Integrated Reasoning

• URL: http://arxiv.org/abs/2605.06326v1
• Date: Thu, 07 May 2026 14:23:21 GMT
• Title: Teaching Thinking Models to Reason with Tools: A Full-Pipeline Recipe for Tool-Integrated Reasoning
• Abstract summary: We show how to inject natural tool-use behavior into a strong thinking model without sacrificing its no-tool reasoning ability.<n>Our recipe yields models that achieve state-of-the-art performance in a wide range of benchmarks among open-source models.
• Score: 59.74608632210439
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Tool-integrated reasoning (TIR) offers a direct way to extend thinking models beyond the limits of text-only reasoning. Paradoxically, we observe that tool-enabled evaluation can degrade reasoning performance even when the strong thinking models make almost no actual tool calls. In this paper, we investigate how to inject natural tool-use behavior into a strong thinking model without sacrificing its no-tool reasoning ability, and present a comprehensive TIR recipe. We highlight that (i) the effectiveness of TIR supervised fine-tuning (SFT) hinges on the learnability of teacher trajectories, which should prioritize problems inherently suited for tool-augmented solutions; (ii) controlling the proportion of tool-use trajectories could mitigate the catastrophic forgetting of text-only reasoning capacity; (iii) optimizing for pass@k and response length instead of training loss could maximize TIR SFT gains while preserving headroom for reinforcement learning (RL) exploration; (iv) a stable RL with verifiable rewards (RLVR) stage, built upon suitable SFT initialization and explicit safeguards against mode collapse, provides a simple yet remarkably effective solution. When applied to Qwen3 thinking models at 4B and 30B scales, our recipe yields models that achieve state-of-the-art performance in a wide range of benchmarks among open-source models, such as 96.7% and 99.2% on AIME 2025 for 4B and 30B, respectively.

Related papers

• Beyond Distribution Sharpening: The Importance of Task Rewards [20.11905685439596]
  We compare distribution sharpening and task-reward-based learning, utilizing RL as a tool to implement both paradigms.<n>Experiments using Llama-3.2-3B-Instruct, Qwen2.5-3B-Instruct and Qwen3-4B-Instruct-2507 on math datasets confirm that sharpening yields limited gains.
  arXiv  Detail & Related papers  (2026-04-17T17:17:55Z)
• In-Context Reinforcement Learning for Tool Use in Large Language Models [68.66653829365187]
  Large language models (LLMs) exhibit strong reasoning abilities, but their performance on complex tasks is often constrained by their internal knowledge.<n>We propose In-Context Reinforcement Learning (ICRL), an RL-only framework that eliminates the need for supervised fine-tuning.<n>ICRL achieves state-of-the-art performance, demonstrating its effectiveness as a scalable, data-efficient alternative to traditional SFT-based pipelines.
  arXiv  Detail & Related papers  (2026-03-09T08:06:18Z)
• One Model to Critique Them All: Rewarding Agentic Tool-Use via Efficient Reasoning [54.580646706013965]
  Reward models (RMs) play a critical role in aligning large language models with human preferences.<n>We introduce ToolRM, a family of lightweight generative RMs tailored for general tool-use scenarios.<n>To build these models, we propose a novel pipeline that constructs pairwise preference data using rule-based scoring and multidimensional sampling.
  arXiv  Detail & Related papers  (2025-10-30T06:08:27Z)
• Demystifying Reinforcement Learning in Agentic Reasoning [90.3737088727791]
  We conduct a comprehensive and systematic investigation to demystify reinforcement learning in agentic reasoning.<n>We highlight our key insights: (i) replacing stitched synthetic trajectories with real end-to-end tool-use trajectories yields a far stronger SFT.<n> Exploration-friendly techniques are crucial for agentic RL, such as clip higher, overlong reward shaping, and maintaining adequate policy entropy could improve the training efficiency.
  arXiv  Detail & Related papers  (2025-10-13T17:57:15Z)
• Nemotron-Research-Tool-N1: Exploring Tool-Using Language Models with Reinforced Reasoning [93.30252692375886]
  Rule-based reinforcement learning can be used to enhance tool-calling in large language models.<n>Tool-N1-7B/14B clearly outperform GPT-4o on several major benchmarks.
  arXiv  Detail & Related papers  (2025-04-25T02:55:21Z)
• Acting Less is Reasoning More! Teaching Model to Act Efficiently [87.28134636548705]
  Tool-integrated reasoning augments large language models with the ability to invoke external tools to solve tasks.<n>Current approaches typically optimize only for final correctness without considering the efficiency or necessity of external tool use.<n>We propose a framework that encourages models to produce accurate answers with minimal tool calls.<n>Our approach reduces tool calls by up to 68.3% and improves tool productivity by up to 215.4%, while maintaining comparable answer accuracy.
  arXiv  Detail & Related papers  (2025-04-21T05:40:05Z)
• ReTool: Reinforcement Learning for Strategic Tool Use in LLMs [27.07998056454784]
  ReTool enhances long-form reasoning with tool-integrated learning.<n>Model achieves 67% accuracy with 400 training steps.<n>Remarkably, ReTool-32B attains 72.5% accuracy in extended settings.
  arXiv  Detail & Related papers  (2025-04-15T18:10:22Z)

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.