Shape of Thought: When Distribution Matters More than Correctness in Reasoning Tasks

• URL: http://arxiv.org/abs/2512.22255v1
• Date: Wed, 24 Dec 2025 07:35:55 GMT
• Title: Shape of Thought: When Distribution Matters More than Correctness in Reasoning Tasks
• Authors: Abhranil Chandra, Ayush Agrawal, Arian Hosseini, Sebastian Fischmeister, Rishabh Agarwal, Navin Goyal, Aaron Courville,
• Abstract summary: We show that a language model's reasoning capabilities can be improved by training on datasets of chain-of-thought traces from more capable models.<n>Experiments show this approach can yield better performance on reasoning tasks than training on human-annotated datasets.
• Score: 24.55929874173401
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We present the surprising finding that a language model's reasoning capabilities can be improved by training on synthetic datasets of chain-of-thought (CoT) traces from more capable models, even when all of those traces lead to an incorrect final answer. Our experiments show this approach can yield better performance on reasoning tasks than training on human-annotated datasets. We hypothesize that two key factors explain this phenomenon: first, the distribution of synthetic data is inherently closer to the language model's own distribution, making it more amenable to learning. Second, these `incorrect' traces are often only partially flawed and contain valid reasoning steps from which the model can learn. To further test the first hypothesis, we use a language model to paraphrase human-annotated traces -- shifting their distribution closer to the model's own distribution -- and show that this improves performance. For the second hypothesis, we introduce increasingly flawed CoT traces and study to what extent models are tolerant to these flaws. We demonstrate our findings across various reasoning domains like math, algorithmic reasoning and code generation using MATH, GSM8K, Countdown and MBPP datasets on various language models ranging from 1.5B to 9B across Qwen, Llama, and Gemma models. Our study shows that curating datasets that are closer to the model's distribution is a critical aspect to consider. We also show that a correct final answer is not always a reliable indicator of a faithful reasoning process.

Related papers

• Mitigating Spurious Correlations Between Question and Answer via Chain-of-Thought Correctness Perception Distillation [25.195244084313114]
  Chain-of-Thought Correctness Perception Distillation (CoPeD) aims to improve the reasoning quality of the student model.<n>CoPeD encourages the student model to predict answers based on correct rationales and revise them when they are incorrect.
  arXiv  Detail & Related papers  (2025-09-06T05:33:17Z)
• A Closer Look at Bias and Chain-of-Thought Faithfulness of Large (Vision) Language Models [58.32070787537946]
  Chain-of-thought (CoT) reasoning enhances performance of large language models.<n>We present the first comprehensive study of CoT faithfulness in large vision-language models.
  arXiv  Detail & Related papers  (2025-05-29T18:55:05Z)
• Beyond Semantics: The Unreasonable Effectiveness of Reasonless Intermediate Tokens [14.78605805191225]
  We investigate how the semantics of intermediate tokens-often anthropomorphized as "thoughts" or reasoning traces-actually influence model performance.<n>We show that despite significant improvements on the solution-only baseline, models trained on entirely correct traces still produce invalid reasoning traces when arriving at correct solutions.
  arXiv  Detail & Related papers  (2025-05-19T23:29:23Z)
• Do Larger Language Models Generalize Better? A Scaling Law for Implicit Reasoning at Pretraining Time [73.22651918134808]
  This work shows counterintuitive effects of model size scaling and provides new insights into the relationship between scaling and reasoning in language models (LMs)<n>We pretrain LMs from scratch on a synthetic implicit multihop reasoning environment designed to replicate the structure and distribution of real-world large-scale knowledge graphs.<n>We then assess the LMs' ability to complete the missing edges in the graph, which requires multi-hop reasoning that can be viewed as a simplification of implicit reasoning during real-world pretraining.
  arXiv  Detail & Related papers  (2025-04-04T17:57:22Z)
• Towards Thinking-Optimal Scaling of Test-Time Compute for LLM Reasoning [108.07030347318624]
  We show that scaling with longer Chain of Thoughts (CoTs) can indeed impair the reasoning performance of Large Language Models (LLMs) in certain domains.<n>We propose a Thinking- Optimal Scaling strategy to teach models to adopt different reasoning efforts for deep thinking.<n>Our self-improvement models built upon Qwen2.5-32B-Instruct outperform other distillation-based 32B o1-like models across various math benchmarks.
  arXiv  Detail & Related papers  (2025-02-25T10:48:05Z)
• Self-supervised Analogical Learning using Language Models [59.64260218737556]
  We propose SAL, a self-supervised analogical learning framework.<n> SAL mimics the human analogy process and trains models to explicitly transfer high-quality symbolic solutions.<n>We show that the resulting models outperform base language models on a wide range of reasoning benchmarks.
  arXiv  Detail & Related papers  (2025-02-03T02:31:26Z)
• Improving the Natural Language Inference robustness to hard dataset by data augmentation and preprocessing [1.7487745673871375]
  Natural Language Inference (NLI) is the task of inferring whether the hypothesis can be justified by the given premise.<n>We propose the data augmentation and preprocessing methods to solve the word overlap, numerical reasoning and length mismatch problems.
  arXiv  Detail & Related papers  (2024-12-10T01:49:23Z)
• Debiasing Multimodal Models via Causal Information Minimization [65.23982806840182]
  We study bias arising from confounders in a causal graph for multimodal data. Robust predictive features contain diverse information that helps a model generalize to out-of-distribution data. We use these features as confounder representations and use them via methods motivated by causal theory to remove bias from models.
  arXiv  Detail & Related papers  (2023-11-28T16:46:14Z)
• Are Data-driven Explanations Robust against Out-of-distribution Data? [18.760475318852375]
  We propose an end-to-end model-agnostic learning framework Distributionally Robust Explanations (DRE) Key idea is to fully utilize the inter-distribution information to provide supervisory signals for the learning of explanations without human annotation. Our results demonstrate that the proposed method significantly improves the model's performance in terms of explanation and prediction robustness against distributional shifts.
  arXiv  Detail & Related papers  (2023-03-29T02:02:08Z)
• A Multi-Level Attention Model for Evidence-Based Fact Checking [58.95413968110558]
  We present a simple model that can be trained on sequence structures. Results on a large-scale dataset for Fact Extraction and VERification show that our model outperforms the graph-based approaches.
  arXiv  Detail & Related papers  (2021-06-02T05:40:12Z)
• Why do classifier accuracies show linear trends under distribution shift? [58.40438263312526]
  accuracies of models on one data distribution are approximately linear functions of the accuracies on another distribution. We assume the probability that two models agree in their predictions is higher than what we can infer from their accuracy levels alone. We show that a linear trend must occur when evaluating models on two distributions unless the size of the distribution shift is large.
  arXiv  Detail & Related papers  (2020-12-31T07:24:30Z)
• CausaLM: Causal Model Explanation Through Counterfactual Language Models [33.29636213961804]
  CausaLM is a framework for producing causal model explanations using counterfactual language representation models. We show that language representation models such as BERT can effectively learn a counterfactual representation for a given concept of interest. A byproduct of our method is a language representation model that is unaffected by the tested concept.
  arXiv  Detail & Related papers  (2020-05-27T15:06:35Z)

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.