API Is Enough: Conformal Prediction for Large Language Models Without Logit-Access

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Su, Jiayuan, Luo, Jing, Wang, Hongwei, Cheng, Lu
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913298162647040
author Su, Jiayuan
Luo, Jing
Wang, Hongwei
Cheng, Lu
author_facet Su, Jiayuan
Luo, Jing
Wang, Hongwei
Cheng, Lu
contents This study aims to address the pervasive challenge of quantifying uncertainty in large language models (LLMs) without logit-access. Conformal Prediction (CP), known for its model-agnostic and distribution-free features, is a desired approach for various LLMs and data distributions. However, existing CP methods for LLMs typically assume access to the logits, which are unavailable for some API-only LLMs. In addition, logits are known to be miscalibrated, potentially leading to degraded CP performance. To tackle these challenges, we introduce a novel CP method that (1) is tailored for API-only LLMs without logit-access; (2) minimizes the size of prediction sets; and (3) ensures a statistical guarantee of the user-defined coverage. The core idea of this approach is to formulate nonconformity measures using both coarse-grained (i.e., sample frequency) and fine-grained uncertainty notions (e.g., semantic similarity). Experimental results on both close-ended and open-ended Question Answering tasks show our approach can mostly outperform the logit-based CP baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2403_01216
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle API Is Enough: Conformal Prediction for Large Language Models Without Logit-Access
Su, Jiayuan
Luo, Jing
Wang, Hongwei
Cheng, Lu
Computation and Language
Artificial Intelligence
Machine Learning
This study aims to address the pervasive challenge of quantifying uncertainty in large language models (LLMs) without logit-access. Conformal Prediction (CP), known for its model-agnostic and distribution-free features, is a desired approach for various LLMs and data distributions. However, existing CP methods for LLMs typically assume access to the logits, which are unavailable for some API-only LLMs. In addition, logits are known to be miscalibrated, potentially leading to degraded CP performance. To tackle these challenges, we introduce a novel CP method that (1) is tailored for API-only LLMs without logit-access; (2) minimizes the size of prediction sets; and (3) ensures a statistical guarantee of the user-defined coverage. The core idea of this approach is to formulate nonconformity measures using both coarse-grained (i.e., sample frequency) and fine-grained uncertainty notions (e.g., semantic similarity). Experimental results on both close-ended and open-ended Question Answering tasks show our approach can mostly outperform the logit-based CP baselines.
title API Is Enough: Conformal Prediction for Large Language Models Without Logit-Access
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2403.01216