paper-with-me

Papers

Social Bias Evaluation for Large Language Models Requires Prompt Variations

2024-07-03 · Rem Hida, Masahiro Kaneko, Naoaki Okazaki

Warning: This paper contains examples of stereotypes and biases. Large Language Models (LLMs) exhibit considerable social biases, and various studies have tried to evaluate and mitigate these biases accurately. Previous studies use downstream tasks as prompts to examine the degree of social biases for evaluation and mitigation. While LLMs' output highly depends on prompts, previous studies evaluating and mitigating bias have often relied on a limited variety of prompts. In this paper, we investigate the sensitivity of LLMs when changing prompt variations (task instruction and prompt, few-shot examples, debias-prompt) by analyzing task performance and social bias of LLMs. Our experimental results reveal that LLMs are highly sensitive to prompts to the extent that the ranking of LLMs fluctuates when comparing models for task performance and social bias. Additionally, we show that LLMs have tradeoffs between performance and social bias caused by the prompts. Less bias from prompt setting may result in reduced performance. Moreover, the ambiguity of instances is one of the reasons for this sensitivity to prompts in advanced LLMs, leading to various outputs. We recommend using diverse prompts, as in this study, to compare the effects of prompts on social bias in LLMs.

📄 PDF Abstract BibTeX arXiv:2407.03129

Code (1)

rem-h4/llm_socialbias_prompts 공식 구현 pytorch

Tasks

Sensitivity

Similar Papers 제목 키워드 기반

KoSBi: A Dataset for Mitigating Social Bias Risks Towards Safer Large Language Model Application

2023-05-28 · Hwaran Lee, Seokhee Hong, Joonsuk Park, Takyoung Kim 외

Large language models (LLMs) learn not only natural text generation abilities but also social biases against different demographic groups from real-world data. This poses a critical risk when deploying LLM-based applicat…

Language ModelingLanguage ModellingLarge Language ModelText Generation

EsBBQ and CaBBQ: The Spanish and Catalan Bias Benchmarks for Question Answering

2025-07-15 · Valle Ruiz-Fernández, Mario Mina, Júlia Falcão, Luis Vasquez-Reina 외 arxiv

Previous literature has largely shown that Large Language Models (LLMs) perpetuate social biases learnt from their pre-training data. Given the notable lack of resources for social bias evaluation in languages other than…

Question Answering

Explicit vs. Implicit: Investigating Social Bias in Large Language Models through Self-Reflection

2025-01-04 · Yachao Zhao, Bo wang, Yan Wang

Large Language Models (LLMs) have been shown to exhibit various biases and stereotypes in their generated content. While extensive research has investigated bias in LLMs, prior work has predominantly focused on explicit …

Social Bias Benchmark for Generation: A Comparison of Generation and QA-Based Evaluations

2025-03-10 · Jiho Jin, Woosung Kang, Junho Myung, Alice Oh

Measuring social bias in large language models (LLMs) is crucial, but existing bias evaluation methods struggle to assess bias in long-form generation. We propose a Bias Benchmark for Generation (BBG), an adaptation of t…

FormMultiple-choiceStory Generation

You reap what you sow: On the Challenges of Bias Evaluation Under Multilingual Settings

2022-05-01 · BigScience (ACL) 2022 5 · Zeerak Talat, Aurélie Névéol, Stella Biderman, Miruna Clinciu 외

Evaluating bias, fairness, and social impact in monolingual language models is a difficult task. This challenge is further compounded when language modeling occurs in a multilingual context. Considering the implication o…

FairnessLanguage ModelingLanguage Modelling