See What LLMs Cannot Answer: A Self-Challenge Framework for Uncovering LLM Weaknesses
Westlake University(西湖大学);University of Cambridge(剑桥大学);Microsoft GenAI(微软生成式人工智能部门);UIUC(伊利诺伊大学厄巴纳-香槟分校);Westlake Institute for Advanced Study(西湖高等研究院)
Yulong Chen, Yang Liu, Jianhao Yan, Xuefeng Bai, Ming Zhong, Yinghao Yang, Ziyi Yang, Chenguang Zhu, Yue Zhang
The impressive performance of Large Language Models (LLMs) has consistently surpassed numerous human-designed benchmarks, presenting new challenges in assessing the shortcomings of LLMs. Designing tasks and finding LLMs' limitations are becoming increasingly important. In this paper, we investigate the question of whether an LLM can discover its own limitations from the errors it makes. To this end, we propose a Self-Challenge evaluation framework with human-in-the-loop. Starting from seed instances that GPT-4 fails to answer, we prompt GPT-4 to summarize error patterns that can be used to generate new instances and incorporate human feedback on them to refine these patterns for generating more challenging data, iteratively. We end up with 8 diverse patterns, such as text manipulation and questions with assumptions. We then build a benchmark, SC-G4, consisting of 1,835 instances generated by GPT-4 using these patterns, with human-annotated gold responses. The SC-G4 serves as a challenging benchmark that allows for a detailed assessment of LLMs' abilities. Our results show that only 44.96\% of instances in SC-G4 can be answered correctly by GPT-4. Interestingly, our pilot study indicates that these error patterns also challenge other LLMs, such as Claude-3 and Llama-3, and cannot be fully resolved through fine-tuning. Our work takes the first step to demonstrate that LLMs can autonomously identify their inherent flaws and provide insights for future dynamic and automatic evaluation.
Uncovering Limitations of Large Language Models in Information Seeking from Tables
Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS)(中国科学院智能信息处理重点实验室);Institute of Computing Technology, CAS(中国科学院计算技术研究所);University of Chinese Academy of Sciences(中国科学院大学);Harbin Engineering University(哈尔滨工程大学);Peng Cheng Laboratory(鹏城实验室)
Identifying the Achilles' Heel: An Iterative Method for Dynamically Uncovering Factual Errors in Large Language Models
Renmin University of China(中国人民大学);The Chinese University of Hong Kong(香港中文大学);The Chinese University of Hong Kong, Shenzhen(香港中文大学深圳校区);Johns Hopkins University(约翰霍普金斯大学);Nanyang Technological University(南洋理工大学);Xiaohongshu Inc.(小红书公司);Tencent Inc.(腾讯公司)
Revealing Weaknesses in Text Watermarking Through Self-Information Rewrite Attacks
University of British Columbia(不列颠哥伦比亚大学);Vector Institute for AI(向量人工智能研究所);University of Cambridge(剑桥大学);Canada CIFAR AI Chair(加拿大CIFAR人工智能主席);NSERC CRC Chair(NSERC加拿大研究主席)
U-Space: Uncovering When and Why Uncertainty Arises in Language Models
Technische Universität Darmstadt(达姆施塔特工业大学);University College London(伦敦大学学院);Universität zu Lübeck(吕贝克大学);Mila – Quebec Artificial Intelligence Institute(米拉-魁北克人工智能研究所);Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学)
LITA: An Efficient LLM-assisted Iterative Topic Augmentation Framework
Department of Biomedical Informatics & Data Science, Yale University(生物医学信息学与数据科学系,耶鲁大学);Department of Information Management, National Sun Yat-sen University(信息管理系,National Sun Yat-sen University)
Are VLMs Seeing or Just Saying? Uncovering the Illusion of Visual Re-examination
University of Southern California(南加州大学);University of California San Diego(加州大学圣地亚哥分校);Carnegie Mellon University(卡内基梅隆大学);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)