CritiCal: Can Critique Help LLM Uncertainty or Confidence Calibration?
By: Qing Zong , Jiayu Liu , Tianshi Zheng and more
Potential Business Impact:
Makes AI tell you when it's unsure.
Accurate confidence calibration in Large Language Models (LLMs) is critical for safe use in high-stakes domains, where clear verbalized confidence enhances user trust. Traditional methods that mimic reference confidence expressions often fail to capture the reasoning needed for accurate confidence assessment. We propose natural language critiques as a solution, ideally suited for confidence calibration, as precise gold confidence labels are hard to obtain and often require multiple generations. This paper studies how natural language critiques can enhance verbalized confidence, addressing: (1) What to critique: uncertainty (question-focused) or confidence (answer-specific)? Analysis shows confidence suits multiple-choice tasks, while uncertainty excels in open-ended scenarios. (2) How to critique: self-critique or critique calibration training? We propose Self-Critique, enabling LLMs to critique and optimize their confidence beyond mere accuracy, and CritiCal, a novel Critique Calibration training method that leverages natural language critiques to improve confidence calibration, moving beyond direct numerical optimization. Experiments show that CritiCal significantly outperforms Self-Critique and other competitive baselines, even surpassing its teacher model, GPT-4o, in complex reasoning tasks. CritiCal also shows robust generalization in out-of-distribution settings, advancing LLM's reliability.
Similar Papers
Comparing Uncertainty Measurement and Mitigation Methods for Large Language Models: A Systematic Review
Computation and Language
Makes AI tell the truth, not make things up.
How to Correctly Report LLM-as-a-Judge Evaluations
Machine Learning (CS)
Fixes computer judge mistakes for fairer tests.
Beyond Accuracy: The Role of Calibration in Self-Improving Large Language Models
Computation and Language
Makes AI more honest about what it knows.