Estimating LLM Uncertainty with Evidence
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913827824599040 |
|---|---|
| author | Ma, Huan Chen, Jingdong Zhou, Joey Tianyi Wang, Guangyu Zhang, Changqing |
| author_facet | Ma, Huan Chen, Jingdong Zhou, Joey Tianyi Wang, Guangyu Zhang, Changqing |
| contents | Over the past few years, Large Language Models (LLMs) have developed rapidly and are widely applied in various domains. However, LLMs face the issue of hallucinations, generating responses that may be unreliable when the models lack relevant knowledge. To be aware of potential hallucinations, uncertainty estimation methods have been introduced, and most of them have confirmed that reliability lies in critical tokens. However, probability-based methods perform poorly in identifying token reliability, limiting their practical utility. In this paper, we reveal that the probability-based method fails to estimate token reliability due to the loss of evidence strength information which is accumulated in the training stage. Therefore, we present Logits-induced token uncertainty (LogTokU), a framework for estimating decoupled token uncertainty in LLMs, enabling real-time uncertainty estimation without requiring multiple sampling processes. We employ evidence modeling to implement LogTokU and use the estimated uncertainty to guide downstream tasks. The experimental results demonstrate that LogTokU has significant effectiveness and promise. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_00290 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Estimating LLM Uncertainty with Evidence Ma, Huan Chen, Jingdong Zhou, Joey Tianyi Wang, Guangyu Zhang, Changqing Computation and Language Artificial Intelligence Over the past few years, Large Language Models (LLMs) have developed rapidly and are widely applied in various domains. However, LLMs face the issue of hallucinations, generating responses that may be unreliable when the models lack relevant knowledge. To be aware of potential hallucinations, uncertainty estimation methods have been introduced, and most of them have confirmed that reliability lies in critical tokens. However, probability-based methods perform poorly in identifying token reliability, limiting their practical utility. In this paper, we reveal that the probability-based method fails to estimate token reliability due to the loss of evidence strength information which is accumulated in the training stage. Therefore, we present Logits-induced token uncertainty (LogTokU), a framework for estimating decoupled token uncertainty in LLMs, enabling real-time uncertainty estimation without requiring multiple sampling processes. We employ evidence modeling to implement LogTokU and use the estimated uncertainty to guide downstream tasks. The experimental results demonstrate that LogTokU has significant effectiveness and promise. |
| title | Estimating LLM Uncertainty with Evidence |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2502.00290 |