B-score: Detecting biases in large language models using response history

• URL: http://arxiv.org/abs/2505.18545v1
• Date: Sat, 24 May 2025 06:23:52 GMT
• Title: B-score: Detecting biases in large language models using response history
• Authors: An Vo, Mohammad Reza Taesiri, Daeyoung Kim, Anh Totti Nguyen,
• Abstract summary: Large language models (LLMs) often exhibit strong biases.<n>We investigate whether LLMs would be able to output less biased answers when allowed to observe their prior answers to the same question.<n>We propose B-score, a novel metric that is effective in detecting biases to Subjective, Random, Easy, and Hard questions.
• Score: 2.944057642865492
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) often exhibit strong biases, e.g, against women or in favor of the number 7. We investigate whether LLMs would be able to output less biased answers when allowed to observe their prior answers to the same question in a multi-turn conversation. To understand which types of questions invite more biased answers, we test LLMs on our proposed set of questions that span 9 topics and belong to three types: (1) Subjective; (2) Random; and (3) Objective. Interestingly, LLMs are able to "de-bias" themselves in a multi-turn conversation in response to questions that seek an Random, unbiased answer. Furthermore, we propose B-score, a novel metric that is effective in detecting biases to Subjective, Random, Easy, and Hard questions. On MMLU, HLE, and CSQA, leveraging B-score substantially improves the verification accuracy of LLM answers (i.e, accepting LLM correct answers and rejecting incorrect ones) compared to using verbalized confidence scores or the frequency of single-turn answers alone. Code and data are available at: https://b-score.github.io.

Related papers

• Do LLMs have a Gender (Entropy) Bias? [3.2225437367979763]
  We define and study entropy bias, which we define as a discrepancy in the amount of information generated by an LLM in response to real questions users have asked.<n>Our analyses suggest that there is no significant bias in LLM responses for men and women at a category level.<n>We suggest a simple debiasing approach that iteratively merges the responses for the two genders to produce a final result.
  arXiv  Detail & Related papers  (2025-05-24T23:06:41Z)
• Right Answer, Wrong Score: Uncovering the Inconsistencies of LLM Evaluation in Multiple-Choice Question Answering [78.89231943329885]
  Multiple-Choice Question Answering (MCQA) is widely used to evaluate Large Language Models (LLMs)<n>We show that multiple factors can significantly impact the reported performance of LLMs.<n>We analyze whether existing answer extraction methods are aligned with human judgment.
  arXiv  Detail & Related papers  (2025-03-19T08:45:03Z)
• Multiple Choice Questions: Reasoning Makes Large Language Models (LLMs) More Self-Confident Even When They Are Wrong [2.8367942280334493]
  We study how the confidence in the answer depends on whether the model has been asked to answer directly or to provide the reasoning before answering.<n>Our hypothesis is that this behavior is due to the reasoning that modifies the probability of the selected answer.
  arXiv  Detail & Related papers  (2025-01-16T10:27:51Z)
• Are LLMs Aware that Some Questions are not Open-ended? [58.93124686141781]
  We study whether Large Language Models are aware that some questions have limited answers and need to respond more deterministically. The lack of question awareness in LLMs leads to two phenomena: (1) too casual to answer non-open-ended questions or (2) too boring to answer open-ended questions.
  arXiv  Detail & Related papers  (2024-10-01T06:07:00Z)
• 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)
• Evaluating Nuanced Bias in Large Language Model Free Response Answers [8.775925011558995]
  We identify several kinds of nuanced bias in free text that cannot be identified by multiple choice tests. We present a semi-automated pipeline for detecting these types of bias by first eliminating answers that can be automatically classified as unbiased.
  arXiv  Detail & Related papers  (2024-07-11T19:58:13Z)
• Open-LLM-Leaderboard: From Multi-choice to Open-style Questions for LLMs Evaluation, Benchmark, and Arena [23.264049073539663]
  Multiple-choice questions (MCQ) are frequently used to assess large language models (LLMs) LLMs may inherently favor certain answer choice IDs, such as A/B/C/D, due to inherent biases of priori unbalanced probabilities. This work aims to tackle these significant difficulties, and establish a new LLM evaluation benchmark through entirely open-style questions.
  arXiv  Detail & Related papers  (2024-06-11T17:59:47Z)
• Don't Just Say "I don't know"! Self-aligning Large Language Models for Responding to Unknown Questions with Explanations [70.6395572287422]
  Self-alignment method is capable of not only refusing to answer but also providing explanation to the unanswerability of unknown questions. We conduct disparity-driven self-curation to select qualified data for fine-tuning the LLM itself for aligning the responses to unknown questions as desired.
  arXiv  Detail & Related papers  (2024-02-23T02:24:36Z)
• Large Language Models Are Not Robust Multiple Choice Selectors [117.72712117510953]
  Multiple choice questions (MCQs) serve as a common yet important task format in the evaluation of large language models (LLMs) This work shows that modern LLMs are vulnerable to option position changes due to their inherent "selection bias" We propose a label-free, inference-time debiasing method, called PriDe, which separates the model's prior bias for option IDs from the overall prediction distribution.
  arXiv  Detail & Related papers  (2023-09-07T17:44:56Z)
• Large Language Models are Better Reasoners with Self-Verification [48.534270563880845]
  Large language models (LLMs) have shown strong reasoning ability in several natural language processing tasks. LLMs with chain of thought (CoT) prompting require multi-step prompting and multi-token prediction, which is highly sensitive to individual mistakes. We propose and prove that LLMs also have similar self-verification abilities.
  arXiv  Detail & Related papers  (2022-12-19T15:51:52Z)

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.