How to Steal Reasoning Without Reasoning Traces

• URL: http://arxiv.org/abs/2603.07267v1
• Date: Sat, 07 Mar 2026 15:50:44 GMT
• Title: How to Steal Reasoning Without Reasoning Traces
• Abstract summary: Many large language models (LLMs) use reasoning to generate responses but do not reveal their full reasoning traces.<n>We introduce trace inversion models that, given only the inputs, answers, and (optionally) reasoning summaries exposed by a target model, generate detailed, synthetic reasoning traces.
• Score: 11.96740610715437
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Many large language models (LLMs) use reasoning to generate responses but do not reveal their full reasoning traces (a.k.a. chains of thought), instead outputting only final answers and brief reasoning summaries. To demonstrate that hiding reasoning traces does not prevent users from "stealing" a model's reasoning capabilities, we introduce trace inversion models that, given only the inputs, answers, and (optionally) reasoning summaries exposed by a target model, generate detailed, synthetic reasoning traces. We show that (1) traces synthesized by trace inversion have high overlap with the ground-truth reasoning traces (when available), and (2) fine-tuning student models on inverted traces substantially improves their reasoning. For example, fine-tuning Qwen-2.5-7B-Instruct on traces inverted from the answers and summaries of GPT-5 mini, a commercial black-box LLM, improves its performance from 56.8% to 77.6% on MATH500 and from 11.7% to 42.3% on JEEBench, compared to fine-tuning on just the answers and summaries.

Related papers

• ThinkRetrieve: Retrieval-Augmented Reasoning Traces for Test-Time Scaling [60.48279351544583]
  ThinkRetrieve is a test-time scaling framework for large reasoning models.<n>It retrieves relevant exemplars at each intermediate step and injects them directly into the thinking trace.<n>It consistently improves accuracy over standard test-time scaling.
  arXiv  Detail & Related papers  (2026-08-11T13:58:07Z)
• Reasoning Traces Shape Outputs but Models Won't Say So [11.928107300132062]
  We introduce Thought Injection, a method that injects synthetic reasoning snippets into a model's think> trace.<n>We find that injected hints reliably alter outputs, confirming that reasoning traces causally shape model behavior.<n>When asked to explain their changed answers, models overwhelmingly refuse to disclose the influence.
  arXiv  Detail & Related papers  (2026-03-21T03:31:36Z)
• Measuring Reasoning Trace Legibility: Can Those Who Understand Teach? [14.267421398274571]
  We evaluate 90k traces from 12 Reasoning Language Models (RLMs) for the quality of their reasoning traces.<n>We find that the reasoning traces of the highest-performing models rank among the lowest for legibility.
  arXiv  Detail & Related papers  (2026-03-20T21:23:29Z)
• Probing the Trajectories of Reasoning Traces in Large Language Models [4.599673637363014]
  We propose a protocol to probe the trajectories of reasoning traces in large language models.<n>We find that accuracy and decision commitment consistently increase as the percentage of provided reasoning tokens grows.<n>We show that trajectory probing provides diagnostics for efficient and safer deployment of reasoning models.
  arXiv  Detail & Related papers  (2026-01-30T16:45:16Z)
• Visual Reasoning Tracer: Object-Level Grounded Reasoning Benchmark [112.46338388724116]
  We introduce the Visual Reasoning Tracer (VRT) task, which requires models to not only localize the target object but also explicitly predict the intermediate objects that form the reasoning path.<n>To advance research in this area, we contribute: (1) VRT-Bench, a human-annotated benchmark for evaluating visual reasoning; (2) a new metric for assessing the quality of reasoning traces; and (3) VRT-80k, a large-scale dataset for reasoning model training.
  arXiv  Detail & Related papers  (2025-12-04T18:55:34Z)
• Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking [50.97239453902612]
  Large Reasoning Models (LRMs) have achieved impressive performance on challenging tasks, yet their deep reasoning often incurs substantial computational costs.<n>Inspired by Evidence Accumulation Models, we find that LRMs have accumulated sufficient information early in reasoning, making further reasoning steps redundant.<n>We propose Just-Enough Thinking (JET), which trains models to proactively terminate unnecessary reasoning.
  arXiv  Detail & Related papers  (2025-09-27T16:25:06Z)
• From Harm to Help: Turning Reasoning In-Context Demos into Assets for Reasoning LMs [58.02809208460186]
  We revisit this paradox using high-quality traces from DeepSeek-R1 as demonstrations.<n>We find that adding more exemplars consistently degrades accuracy, even when demonstrations are optimal.<n>We introduce Insight-to-solve (I2S), a sequential test-time procedure that turns demonstrations into explicit, reusable insights.
  arXiv  Detail & Related papers  (2025-09-27T08:59:31Z)
• FlashThink: An Early Exit Method For Efficient Reasoning [2.1448740411847593]
  Large Language Models (LLMs) have shown impressive performance in reasoning tasks.<n>LLMs tend to generate excessively long reasoning content, leading to significant computational overhead.<n>We introduce a verification model that identifies the exact moment when the model can stop reasoning and still provide the correct answer.
  arXiv  Detail & Related papers  (2025-05-20T05:28:21Z)
• Retro-Search: Exploring Untaken Paths for Deeper and Efficient Reasoning [84.2749507577386]
  We introduce Retro-Search, an MCTS-inspired search algorithm, for distilling higher quality reasoning paths from large models.<n>Retro-Search retrospectively revises reasoning paths to discover better, yet shorter traces, which can lead to student models with enhanced reasoning capabilities.<n>Our approach can enable two use cases: self-improvement, where models are fine-tuned on their own Retro-Search-ed traces, and weak-to-strong improvement.
  arXiv  Detail & Related papers  (2025-04-06T06:23:27Z)
• Benchmarking Reasoning Robustness in Large Language Models [76.79744000300363]
  We find significant performance degradation on novel or incomplete data.<n>These findings highlight the reliance on recall over rigorous logical inference.<n>This paper introduces a novel benchmark, termed as Math-RoB, that exploits hallucinations triggered by missing information to expose reasoning gaps.
  arXiv  Detail & Related papers  (2025-03-06T15:36:06Z)

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.