中文
相关论文

相关论文: Evaluating and Mitigating LLM-as-a-judge Bias in C…

200 篇论文

Over the last year, Large Language Models (LLMs) like ChatGPT have become widely available and have exhibited fairness issues similar to those in previous machine learning systems. Current research is primarily focused on analyzing and…

机器学习 · 计算机科学 2024-04-04 Anna Kruspe

Large Language Models (LLMs) inherit explicit and implicit biases from their training datasets. Identifying and mitigating biases in LLMs is crucial to ensure fair outputs, as they can perpetuate harmful stereotypes and misinformation. This…

机器学习 · 计算机科学 2025-11-19 Fatima Kazi , Alex Young , Yash Inani , Setareh Rafatirad

Large Language Model (LLM) based judges form the underpinnings of key safety evaluation processes such as offline benchmarking, automated red-teaming, and online guardrailing. This widespread requirement raises the crucial question: can we…

机器学习 · 计算机科学 2025-03-07 Francisco Eiras , Eliott Zemour , Eric Lin , Vaikkunth Mugunthan

Generative AI technologies, particularly Large Language Models (LLMs), have transformed information management systems but introduced substantial biases that can compromise their effectiveness in informing business decision-making. This…

计算机与社会 · 计算机科学 2025-02-18 Xiahua Wei , Naveen Kumar , Han Zhang

Large language models (LLMs) offer significant potential as tools to support an expanding range of decision-making tasks. Given their training on human (created) data, LLMs have been shown to inherit societal biases against protected…

人工智能 · 计算机科学 2024-10-07 Jessica Echterhoff , Yao Liu , Abeer Alessa , Julian McAuley , Zexue He

The adoption of generative AI technologies is swiftly expanding. Services employing both linguistic and mul-timodal models are evolving, offering users increasingly precise responses. Consequently, human reliance on these technologies is…

计算机与社会 · 计算机科学 2023-11-17 Jaeyoun You , Bongwon Suh

In this paper, we uncover a systematic bias in the evaluation paradigm of adopting large language models~(LLMs), e.g., GPT-4, as a referee to score and compare the quality of responses generated by candidate models. We find that the quality…

计算与语言 · 计算机科学 2023-08-31 Peiyi Wang , Lei Li , Liang Chen , Zefan Cai , Dawei Zhu , Binghuai Lin , Yunbo Cao , Qi Liu , Tianyu Liu , Zhifang Sui

Large Language Models are increasingly used as judges to evaluate code artifacts when exhaustive human review or executable test coverage is unavailable. LLM-judge is increasingly relevant in agentic software engineering workflows, where it…

软件工程 · 计算机科学 2026-04-21 Zixiao Zhao , Amirreza Esmaeili , Fatemeh Fard

Large language models (LLMs) are increasingly used as automatic evaluators of generative AI outputs, a paradigm often referred to as "LLM-as-a-judge." In practice, LLM judges are imperfect predictions for the underlying truth and can…

机器学习 · 计算机科学 2026-01-12 Yiqun T Chen , Sizhu Lu , Sijia Li , Moran Guo , Shengyi Li

We present a large-scale evaluation of 30 cognitive biases in 20 state-of-the-art large language models (LLMs) under various decision-making scenarios. Our contributions include a novel general-purpose test framework for reliable and…

计算与语言 · 计算机科学 2025-11-04 Simon Malberg , Roman Poletukhin , Carolin M. Schuster , Georg Groh

Large Language Models (LLMs) are increasingly relied upon to evaluate text outputs of other LLMs, thereby influencing leaderboards and development decisions. However, concerns persist over the accuracy of these assessments and the potential…

计算与语言 · 计算机科学 2024-11-27 Sumanth Doddapaneni , Mohammed Safi Ur Rahman Khan , Sshubam Verma , Mitesh M. Khapra

In this study, we investigate the capabilities and inherent biases of advanced large language models (LLMs) such as GPT-3.5 and GPT-4 in the context of debate evaluation. We discover that LLM's performance exceeds humans and surpasses the…

计算与语言 · 计算机科学 2024-06-05 Xinyi Liu , Pinxin Liu , Hangfeng He

Automatic evaluation using multimodal large language models (MLLMs), commonly referred to as MLLM-as-a-Judge, has been widely used to measure model performance. If such MLLM-as-a-Judge methods were biased, they could distort model…

计算机视觉与模式识别 · 计算机科学 2026-04-14 Shuitsu Koyama , Yuiga Wada , Daichi Yashima , Komei Sugiura

Identifying logical errors in complex, incomplete or even contradictory and overall heterogeneous data like students' experimentation protocols is challenging. Recognizing the limitations of current evaluation methods, we investigate the…

人工智能 · 计算机科学 2024-09-20 Arne Bewersdorff , Kathrin Seßler , Armin Baur , Enkelejda Kasneci , Claudia Nerdel

Large language models are increasingly used as judges (LLM-as-a-judge) to evaluate model outputs at scale, but their assessments often diverge systematically from human judgments. We present Bridge, a unified statistical framework that…

机器学习 · 计算机科学 2025-12-03 Felipe Maia Polo , Xinhe Wang , Mikhail Yurochkin , Gongjun Xu , Moulinath Banerjee , Yuekai Sun

Human relevance assessment is time-consuming and cognitively intensive, limiting the scalability of Information Retrieval evaluation. This has led to growing interest in using large language models (LLMs) as proxies for human judges.…

信息检索 · 计算机科学 2026-04-28 Chuting Yu , Hang Li , Guido Zuccon , Joel Mackenzie , Teerapong Leelanupab

Large Language Models (LLMs) have shown powerful performance and development prospects and are widely deployed in the real world. However, LLMs can capture social biases from unprocessed training data and propagate the biases to downstream…

计算与语言 · 计算机科学 2024-02-22 Yingji Li , Mengnan Du , Rui Song , Xin Wang , Ying Wang

When Artificial Intelligence (AI) is used to replace consumers (e.g., synthetic data), it is often assumed that AI emulates established consumers, and more generally human behaviors. Ten experiments with Large Language Models (LLMs)…

Large Language Models (LLMs) are commonly used as evaluators in various applications, but the reliability of the outcomes remains a challenge. One such challenge is using LLMs-as-judges for direct assessment, i.e., assigning scores from a…

计算与语言 · 计算机科学 2026-04-09 Yoshinari Fujinuma

Large Language Models (LLMs) have revolutionized artificial intelligence, demonstrating remarkable computational power and linguistic capabilities. However, these models are inherently prone to various biases stemming from their training…

计算与语言 · 计算机科学 2025-02-14 Riccardo Cantini , Giada Cosenza , Alessio Orsino , Domenico Talia