Related papers: Revisiting Uncertainty Quantification Evaluation in Language Models: Spurious Interactions with Response Length Bias Results

Revisiting Uncertainty Quantification Evaluation in Language Models: Spurious Interactions with Response Length Bias Results

URL: http://arxiv.org/abs/2504.13677v1
Date: Fri, 18 Apr 2025 13:13:42 GMT
Title: Revisiting Uncertainty Quantification Evaluation in Language Models: Spurious Interactions with Response Length Bias Results
Authors: Andrea Santilli, Adam Golinski, Michael Kirchhof, Federico Danieli, Arno Blaas, Miao Xiong, Luca Zappella, Sinead Williamson,
Abstract summary: We show that commonly used correctness functions bias UQ evaluations by inflating the performance of certain UQ methods.<n>We evaluate 7 correctness functions -- from lexical-based and embedding-based metrics to LLM-as-a-judge approaches.
Score: 10.551985027162576
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Abstract: Uncertainty Quantification (UQ) in Language Models (LMs) is crucial for improving their safety and reliability. Evaluations often use performance metrics like AUROC to assess how well UQ methods (e.g., negative sequence probabilities) correlate with task correctness functions (e.g., ROUGE-L). In this paper, we show that commonly used correctness functions bias UQ evaluations by inflating the performance of certain UQ methods. We evaluate 7 correctness functions -- from lexical-based and embedding-based metrics to LLM-as-a-judge approaches -- across 4 datasets x 4 models x 6 UQ methods. Our analysis reveals that length biases in the errors of these correctness functions distort UQ assessments by interacting with length biases in UQ methods. We identify LLM-as-a-judge approaches as among the least length-biased choices and hence a potential solution to mitigate these biases.

Related papers

Right Answer, Wrong Score: Uncovering the Inconsistencies of LLM Evaluation in Multiple-Choice Question Answering [78.89231943329885]
One of the most widely used tasks to evaluate Large Language Models (LLMs) is Multiple-Choice Question Answering (MCQA) In this work, we shed light on the inconsistencies of MCQA evaluation strategies, which can lead to inaccurate and misleading model comparisons.
arXiv Detail & Related papers (2025-03-19T08:45:03Z)
CoT-UQ: Improving Response-wise Uncertainty Quantification in LLMs with Chain-of-Thought [10.166370877826486]
Large language models (LLMs) excel in many tasks but struggle to accurately quantify uncertainty in their generated responses.<n>Existing uncertainty quantification (UQ) methods for LLMs are primarily prompt-wise rather than response-wise, which incurs high computational costs.<n>We propose CoT-UQ, a response-wise UQ framework that integrates LLMs' inherent reasoning capabilities through Chain-of-Thought (CoT) into the UQ process.
arXiv Detail & Related papers (2025-02-24T14:48:06Z)
Token-Level Density-Based Uncertainty Quantification Methods for Eliciting Truthfulness of Large Language Models [76.17975723711886]
Uncertainty quantification (UQ) is a prominent approach for eliciting truthful answers from large language models (LLMs)<n>In this work, we adapt Mahalanobis Distance (MD) - a well-established UQ technique in classification tasks - for text generation.<n>Our method extracts token embeddings from multiple layers of LLMs, computes MD scores for each token, and uses linear regression trained on these features to provide robust uncertainty scores.
arXiv Detail & Related papers (2025-02-20T10:25:13Z)
LINKAGE: Listwise Ranking among Varied-Quality References for Non-Factoid QA Evaluation via LLMs [61.57691505683534]
Non-Factoid (NF) Question Answering (QA) is challenging to evaluate due to diverse potential answers and no objective criterion. Large Language Models (LLMs) have been resorted to for NFQA evaluation due to their compelling performance on various NLP tasks. We propose a novel listwise NFQA evaluation approach, that utilizes LLMs to rank candidate answers in a list of reference answers sorted by descending quality.
arXiv Detail & Related papers (2024-09-23T06:42:21Z)
Benchmarking Uncertainty Quantification Methods for Large Language Models with LM-Polygraph [83.90988015005934]
Uncertainty quantification is a key element of machine learning applications.<n>We introduce a novel benchmark that implements a collection of state-of-the-art UQ baselines.<n>We conduct a large-scale empirical investigation of UQ and normalization techniques across eleven tasks, identifying the most effective approaches.
arXiv Detail & Related papers (2024-06-21T20:06:31Z)
Evaluating Generative Language Models in Information Extraction as Subjective Question Correction [49.729908337372436]
We propose a new evaluation method, SQC-Score. Inspired by the principles in subjective question correction, we propose a new evaluation method, SQC-Score. Results on three information extraction tasks show that SQC-Score is more preferred by human annotators than the baseline metrics.
arXiv Detail & Related papers (2024-04-04T15:36:53Z)
LUQ: Long-text Uncertainty Quantification for LLMs [29.987010627250527]
Large Language Models (LLMs) are prone to generate nonfactual content. Uncertainty Quantification (UQ) is pivotal in enhancing our understanding of a model's confidence on its generation. We propose textscLuq-Ensemble, a method that ensembles responses from multiple models and selects the response with the lowest uncertainty.
arXiv Detail & Related papers (2024-03-29T16:49:24Z)
Towards Clear Expectations for Uncertainty Estimation [64.20262246029286]
Uncertainty Quantification (UQ) is crucial to achieve trustworthy Machine Learning (ML) Most UQ methods suffer from disparate and inconsistent evaluation protocols. This opinion paper offers a new perspective by specifying those requirements through five downstream tasks.
arXiv Detail & Related papers (2022-07-27T07:50:57Z)

This list is automatically generated from the titles and abstracts of the papers in this site.