When Linguistic and Internal Confidence Diverge in Large Language Models

• URL: http://arxiv.org/abs/2608.28382v2
• Date: Fri, 04 Sep 2026 05:39:21 GMT
• Title: When Linguistic and Internal Confidence Diverge in Large Language Models
• Abstract summary: It is unclear whether large language models (LLMs) track their internal confidence.<n>For classification, we compare linguistic confidence with logits-based confidence along three axes: association, magnitude agreement and calibration.<n>For generation, we test whether linguistic confidence tracks semantic-entropy-based uncertainty.
• Score: 36.57105743509995
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Users often ask large language models (LLMs) to report how confident they are, but it is unclear whether such linguistic confidence tracks the model's internal confidence. We study this question across 8 classification tasks, 2 generation tasks and 30 models from three families. For classification, we compare linguistic confidence with logits-based confidence along three axes: association, magnitude agreement and calibration. For generation, we test whether linguistic confidence tracks semantic-entropy-based uncertainty. The axes frequently diverge. Instance-level association is weak on average, although it improves on easier items and for stronger base models. Instruction-tuned models often report higher confidence and sometimes show higher association, but they also have larger confidence gaps and worse calibration. Prompt design mostly changes the distribution of reported confidence. Attitude cues inflate confidence without improving alignment, while score exemplars can preserve rank-order signal when they avoid collapsed confidence values. Regression analyses show that distributional properties of confidence scores explain much of the observed alignment pattern, with model metadata playing a smaller role after controls. These results support a lossy-channel view of linguistic confidence. A more dispersed verbal confidence distribution can carry useful rank information, but it does not make the scores calibrated. Linguistic confidence should therefore be evaluated with multi-axis diagnostics before being used in downstream reliability pipelines.

Related papers

• When Confidence Signals Disagree: Local and Global Confidence in Autoregressive Language Models [0.0]
  We investigate whether different confidence readouts are empirically interchangeable in an autoregressive language model.<n>Across MMLU and ARC Challenge, the two signals are weakly correlated and differ substantially in their association with correctness.<n>Larger local--global confidence gaps are associated with higher answer entropy, more distinct sampled answers, and lower modal-answer concentration.
  arXiv  Detail & Related papers  (2026-09-15T10:08:48Z)
• Speaking in Self-Assessing Tongues: On the Verbalized Confidence of LLMs in Machine Translation [20.39535344813245]
  Unsupervised approaches based on internal signals like predicted probabilities can be misleading because they reflect certainty among alternatives rather than correctness.<n>We devise five verbalized methods of extracting an LLM's per-token confidence without those shortcomings and compare their reliability with that of the model's internal signals of certainty.<n>We find little to no correlation between internal and verbalized methods.
  arXiv  Detail & Related papers  (2026-06-15T19:27:47Z)
• Confidence Geometry Reveals Trace-Level Correctness in Large Language Model Reasoning [5.094079537924077]
  We show that confidence trajectories separate correct from incorrect reasoning traces.<n> correctness-related information is enriched in the tail of reasoning.<n>We propose NeuralConf, a lightweight estimator that learns from confidence trajectories for correctness evaluation.
  arXiv  Detail & Related papers  (2026-05-16T05:57:00Z)
• Identifying Influential N-grams in Confidence Calibration via Regression Analysis [53.95767806659635]
  We identify what linguistic expressions are related to confidence by applying the regression method.<n>We show that large language models (LLMs) remain overconfident when reasoning is involved.<n>We reveal that confidence calibration is possible by simply suppressing those overconfident expressions without drops in performance.
  arXiv  Detail & Related papers  (2026-04-07T11:58:49Z)
• Self-Aware Knowledge Probing: Evaluating Language Models' Relational Knowledge through Confidence Calibration [10.249145599960748]
  Existing knowledge probes evaluate model capabilities through metrics like prediction accuracy and precision.<n>We propose a novel calibration probing framework for relational knowledge, covering three modalities of model confidence.<n>Our analysis reveals that most models, especially those pre-trained with the masking objective, are overconfident.
  arXiv  Detail & Related papers  (2026-01-26T19:11:48Z)
• Can Large Language Models Express Uncertainty Like Human? [71.27418419522884]
  We release the first diverse, large-scale dataset of hedging expressions with human-annotated confidence scores.<n>We conduct the first systematic study of linguistic confidence across modern large language models.
  arXiv  Detail & Related papers  (2025-09-29T02:34:30Z)
• ConfTuner: Training Large Language Models to Express Their Confidence Verbally [58.63318088243125]
  Large Language Models (LLMs) are increasingly deployed in high-stakes domains such as science, law, and healthcare.<n>LLMs are often observed to generate incorrect answers with high confidence, a phenomenon known as "overconfidence"
  arXiv  Detail & Related papers  (2025-08-26T09:25:32Z)
• SteerConf: Steering LLMs for Confidence Elicitation [11.872504642312705]
  Large Language Models (LLMs) exhibit impressive performance across diverse domains but often suffer from overconfidence.<n>We propose SteerConf, a novel framework that systematically steers LLMs' confidence scores to improve their calibration and reliability.
  arXiv  Detail & Related papers  (2025-03-04T18:40:49Z)
• Language Models Prefer What They Know: Relative Confidence Estimation via Confidence Preferences [62.52739672949452]
  Language models (LMs) should provide reliable confidence estimates to help users detect mistakes in their outputs and defer to human experts when necessary.<n>We propose relative confidence estimation, where we match up questions against each other and ask the model to make relative judgments of confidence.<n>Treating each question as a "player" in a series of matchups against other questions and the model's preferences as match outcomes, we can use rank aggregation methods like Elo rating and Bradley-Terry to translate the model's confidence preferences into confidence scores.
  arXiv  Detail & Related papers  (2025-02-03T07:43:27Z)
• Revisiting Confidence Estimation: Towards Reliable Failure Prediction [53.79160907725975]
  We find a general, widely existing but actually-neglected phenomenon that most confidence estimation methods are harmful for detecting misclassification errors. We propose to enlarge the confidence gap by finding flat minima, which yields state-of-the-art failure prediction performance.
  arXiv  Detail & Related papers  (2024-03-05T11:44:14Z)

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.