Research

Paper

AI LLM March 17, 2026

The Cost of Reasoning: Chain-of-Thought Induces Overconfidence in Vision-Language Models

Authors

Robert Welch, Emir Konuk, Kevin Smith

Abstract

Vision-language models (VLMs) are increasingly deployed in high-stakes settings where reliable uncertainty quantification (UQ) is as important as predictive accuracy. Extended reasoning via chain-of-thought (CoT) prompting or reasoning-trained models has become ubiquitous in modern VLM pipelines, yet its effect on UQ reliability remains poorly understood. We show that reasoning consistently degrades the quality of most uncertainty estimates, even when it improves task accuracy. We identify implicit answer conditioning as the primary mechanism: as reasoning traces converge on a conclusion before the final answer is generated, token probabilities increasingly reflect consistency with the model's own reasoning trace rather than uncertainty about correctness. In effect, the model becomes overconfident in its answer. In contrast, agreement-based consistency remains robust and often improves under reasoning, making it a practical choice for uncertainty estimation in reasoning-enabled VLMs.

Metadata

arXiv ID: 2603.16728
Provider: ARXIV
Primary Category: cs.LG
Published: 2026-03-17
Fetched: 2026-03-18 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.16728v1</id>\n    <title>The Cost of Reasoning: Chain-of-Thought Induces Overconfidence in Vision-Language Models</title>\n    <updated>2026-03-17T16:12:06Z</updated>\n    <link href='https://arxiv.org/abs/2603.16728v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.16728v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Vision-language models (VLMs) are increasingly deployed in high-stakes settings where reliable uncertainty quantification (UQ) is as important as predictive accuracy. Extended reasoning via chain-of-thought (CoT) prompting or reasoning-trained models has become ubiquitous in modern VLM pipelines, yet its effect on UQ reliability remains poorly understood. We show that reasoning consistently degrades the quality of most uncertainty estimates, even when it improves task accuracy. We identify implicit answer conditioning as the primary mechanism: as reasoning traces converge on a conclusion before the final answer is generated, token probabilities increasingly reflect consistency with the model's own reasoning trace rather than uncertainty about correctness. In effect, the model becomes overconfident in its answer. In contrast, agreement-based consistency remains robust and often improves under reasoning, making it a practical choice for uncertainty estimation in reasoning-enabled VLMs.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <published>2026-03-17T16:12:06Z</published>\n    <arxiv:primary_category term='cs.LG'/>\n    <author>\n      <name>Robert Welch</name>\n    </author>\n    <author>\n      <name>Emir Konuk</name>\n    </author>\n    <author>\n      <name>Kevin Smith</name>\n    </author>\n  </entry>"
}