Skip to content

Why Do Reasoning Models Lose Coverage? The Role of Data and Forks in the Road

May 2026 · arXiv.org · Vol abs/2605.17026 · 1 citation · 43 references
Computer Science

TL;DR

This paper investigates the cause of reasoning shrinkage under SFT-based post-training and identifies data-centric factors as a key driver of shrinkage in reasoning models and highlights diversity-aware designs as an effective lever for controlling it.

Abstract

Recent progress in large language models has led to the emergence of reasoning models, which have shown strong performance on complex tasks through specialized fine-tuning procedures. While these methods reliably improve pass@1 accuracy, prior works have observed that they show a coverage shrinkage behavior, where pass@k degrades relative to the base model. In this paper, we investigate the cause of reasoning shrinkage under SFT-based post-training. We hypothesize that this behavior is driven by properties of the fine-tuning data, specifically related to decision points or"forks in the road"scenarios where model encounters indecipherable patterns with multiple valid reasoning paths. To test this hypothesis, we design controlled case studies that simulate such decision-point settings, spanning indecipherable nodes in graph branching, and reasoning modes. By tracking post-training dynamics in these settings, we find that the shrinkage phenomenon is tightly correlated with the prevalence of decision-point scenarios in the training data. We also demonstrate that this shrinkage behavior can be partially mitigated through targeted data synthesis design of decision-points and a more systematic diversity-encouraging decoding mechanism. Our findings identify data-centric factors as a key driver of shrinkage in reasoning models and highlight diversity-aware designs as an effective lever for controlling it. (Data and code for reproducing our experiments are available at https://github.com/psunlpgroup/reasoning_forks )

View source

Similar papers

#natural language process... Preprint Sep 2026

Revisiting Complete Reasoning Traces for Post-Training

It is found that full trajectories provide only limited benefit, while partial trajectories are effective even under heavy truncation, and training LLMs using endpoints leads to consistent changes in reasoning behavior, and that it also benefits post-training methods based on reinforcement learning or on-policy distill...

Jaehui Hwang, Sangdoo Yun, Byeongho Heo et al. · 0 citations
#artificial intelligence Preprint Sep 2026

OBC-Prune: Outcome-Based Calibration for Large Reasoning Model Pruning

Large reasoning models (LRMs) generate long chain-of-thought traces before answering, creating significant inference overhead. Pruning can reduce this cost, but its effectiveness depends on the calibration data used to estimate parameter importance. Recent work calibrates on the model's own rollouts instead of generic...

Ha Lan Nguyen, Huy Hoang Son Tran, Trac-Duy Tran et al. · 1 citation
Aug 2026

More Correct Mass, Worse Answers: Why Power Sampling Can Fail and How to Fix It

A deformation-controlled, support-preserving Power target is replaced with a deformation-controlled, support-preserving Power target that calibrates sharpening across problems while limiting the suppression of moderate-probability paths to solve a striking paradox.

Hao-Sen Yang, Jia-Xing Sun, Xiu-Jun Ma · 2 citations
Conference Open access Sep 2026

Sprint or Delve: A Distribution-Aware Approach to Efficient Reasoning

The Powered Length Penalty (PLP) is proposed, an adaptive regularizer that penalizes redundancy in short sequences while gradually reducing penalties for longer sequences, preserving deep reasoning.

Ze-Hui Ling, De-Shu Chen, Hong-Wei Zhang et al. · 0 citations

The Geometry of Logic: Stratification Induces Semantic Structure and Robust Reasoning

STRAT is introduced, which partitions the residual stream into orthogonal Data and Type subspaces and uses Type-based attention and gating to govern Data transformations and in arithmetic, STRAT reduces median OOD error 35-fold relative to a Transformer baseline.

Cristina V. Lopes, Yuan-Gang Li, Justin Tian Jin Chen et al. · 0 citations

When Confidence Rises Too Early: Detecting Shortcut Reasoning via Premature Answer Commitment

The Distributional Answer Commitment Score (DACS) is proposed, a distributional confidence estimator that measures the entropy of the model's probability distribution over answer commitment at each reasoning step to capture how concentrated the model's answer belief is without requiring ground-truth answers or task-spe...

Zhao-Han Zhang, Jun-Jie Liu, Chengzhengxu Li et al. · 0 citations

Related blog posts

MIT News · Artificial Intelligence Nov 17, 2026

Learning the language of machine learning

By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.

Microsoft Research Blog Nov 16, 2026

TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient

AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.