Stumbling Blocks: Stress Testing the Robustness of Machine-Generated Text Detectors Under Attacks
Paul G. Allen School of Computer Science & Engineering, University of Washington(华盛顿大学保罗·G·艾伦计算机科学与工程学院);Xi’an Jiaotong University(西安交通大学);Johns Hopkins University(约翰斯·霍普金斯大学);Peking University(北京大学)
The widespread use of large language models (LLMs) is increasing the demand for methods that detect machine-generated text to prevent misuse. The goal of our study is to stress test the detectors' robustness to malicious attacks under realistic scenarios. We comprehensively study the robustness of popular machine-generated text detectors under attacks from diverse categories: editing, paraphrasing, prompting, and co-generating. Our attacks assume limited access to the generator LLMs, and we compare the performance of detectors on different attacks under different budget levels. Our experiments reveal that almost none of the existing detectors remain robust under all the attacks, and all detectors exhibit different loopholes. Averaging all detectors, the performance drops by 35% across all attacks. Further, we investigate the reasons behind these defects and propose initial out-of-the-box patches to improve robustness.
Clever Hans or Neural Theory of Mind? Stress Testing Social Reasoning in Large Language Models
Bar-Ilan University(巴伊兰大学);University of British Columbia(不列颠哥伦比亚大学);Vector Institute for AI(Vector人工智能研究所);Carnegie Mellon University(卡内基梅隆大学);Allen Institute for Artificial Intelligence(艾伦人工智能研究所);University of Washington(华盛顿大学)
FANToM: A Benchmark for Stress-testing Machine Theory of Mind in Interactions
Allen Institute for Artificial Intelligence(艾伦人工智能研究所);University of Washington(华盛顿大学);Carnegie Mellon University(卡内基梅隆大学);Seoul National University(首尔大学)
Same-Number Citation Swaps: Stress-Testing Jev as a Financial Evidence Judge
Sofia University(索菲亚大学);Indiana University(印第安纳大学);University of California, San Diego(加州大学圣迭戈分校);Northeastern University(东北大学);Michigan State University(密歇根州立大学)
Stress-Testing Multimodal Foundation Models for Crystallographic Reasoning
Dept. of Electrical and Computer Engineering, Texas A&M University(电气与计算机工程系,德克萨斯A&M大学);College of Science and Engineering, Hamad Bin Khalifa University(科学与工程学院,哈马德·本·卡西姆大学);Dept. of Electrical and Computer Engineering, Texas A&M University at Qatar(电气与计算机工程系,德克萨斯A&M大学(卡塔尔));Dept. of Prosthetics and Orthotics, Ankara University(假肢与矫形系,安卡拉大学)
Contextual StereoSet: Stress-Testing Bias Alignment Robustness in Large Language Models
Indian Institute of Information Technology, Allahabad (IIITA)(印度信息与技术研究所(Allahabad));National Institute of Electronics and Information Technology (NIELIT)(国家电子与信息技术研究所)