Guardado en:
Detalles Bibliográficos
Autores principales: Zhang, Glenn, Mayowa, Treasure, Fan, Jason, Fu, Yicheng, Sandoval, Aaron, O'Brien, Sean, Zhu, Kevin
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2512.11998
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911316568965120
author Zhang, Glenn
Mayowa, Treasure
Fan, Jason
Fu, Yicheng
Sandoval, Aaron
O'Brien, Sean
Zhu, Kevin
author_facet Zhang, Glenn
Mayowa, Treasure
Fan, Jason
Fu, Yicheng
Sandoval, Aaron
O'Brien, Sean
Zhu, Kevin
contents Producing trustworthy and reliable Large Language Models (LLMs) has become increasingly important as their usage becomes more widespread. Calibration seeks to achieve this by improving the alignment between the model's confidence and the actual likelihood of its responses being correct or desirable. However, it has been observed that the internal confidence of a model, derived from token probabilities, is not well aligned with its verbalized confidence, leading to misleading results with different calibration methods. In this paper, we propose Direct Confidence Alignment (DCA), a method using Direct Preference Optimization to align an LLM's verbalized confidence with its internal confidence rather than ground-truth accuracy, enhancing model transparency and reliability by ensuring closer alignment between the two confidence measures. We evaluate DCA across multiple open-weight LLMs on a wide range of datasets. To further assess this alignment, we also introduce three new calibration error-based metrics. Our results show that DCA improves alignment metrics on certain model architectures, reducing inconsistencies in a model's confidence expression. However, we also show that it can be ineffective on others, highlighting the need for more model-aware approaches in the pursuit of more interpretable and trustworthy LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2512_11998
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Direct Confidence Alignment: Aligning Verbalized Confidence with Internal Confidence In Large Language Models
Zhang, Glenn
Mayowa, Treasure
Fan, Jason
Fu, Yicheng
Sandoval, Aaron
O'Brien, Sean
Zhu, Kevin
Computation and Language
Producing trustworthy and reliable Large Language Models (LLMs) has become increasingly important as their usage becomes more widespread. Calibration seeks to achieve this by improving the alignment between the model's confidence and the actual likelihood of its responses being correct or desirable. However, it has been observed that the internal confidence of a model, derived from token probabilities, is not well aligned with its verbalized confidence, leading to misleading results with different calibration methods. In this paper, we propose Direct Confidence Alignment (DCA), a method using Direct Preference Optimization to align an LLM's verbalized confidence with its internal confidence rather than ground-truth accuracy, enhancing model transparency and reliability by ensuring closer alignment between the two confidence measures. We evaluate DCA across multiple open-weight LLMs on a wide range of datasets. To further assess this alignment, we also introduce three new calibration error-based metrics. Our results show that DCA improves alignment metrics on certain model architectures, reducing inconsistencies in a model's confidence expression. However, we also show that it can be ineffective on others, highlighting the need for more model-aware approaches in the pursuit of more interpretable and trustworthy LLMs.
title Direct Confidence Alignment: Aligning Verbalized Confidence with Internal Confidence In Large Language Models
topic Computation and Language
url https://arxiv.org/abs/2512.11998