Large language models are widespread, with their performance on benchmarks frequently guiding user preferences for one model over another. However, the vast amount of data these models are trained on can inadvertently lead to contamination with public benchmarks, thus compromising performance measurements. While recently developed contamination detection methods try to address this issue, they overlook the possibility of deliberate contamination by malicious model providers aiming to evade detection. We argue that this setting is of crucial importance as it casts doubt on the reliability of public benchmarks. To more rigorously study this issue, we propose a categorization of both model providers and contamination detection methods. This reveals vulnerabilities in existing methods that we exploit with EAL, a simple yet effective contamination technique that significantly inflates benchmark performance while completely evading current detection methods.
LogProber: Disentangling confidence from contamination in LLM responses
LNC2, INSERM, Paris, France(LNC2,法国国家医学研究院,巴黎);DEC, ENS, PSL, Paris, France(DEC,巴黎高等师范学院,巴黎);Flowers AI & CogSci Lab Centre Inria de L’Université de Bordeaux(AI与认知科学实验室,Inria-波尔多大学研究中心)
Evaluating Personality Traits in Large Language Models: Insights from Psychological Questionnaires
School of Physics, Mathematics and Computing(物理、数学与计算学院);University of Western Australia(西澳大学);School of Computing(计算机学院);School of Psychological Sciences(心理科学学院)
Deep Learning for Network Anomaly Detection under Data Contamination: Evaluating Robustness and Mitigating Performance Degradation
Groupe de Recherche Interdisciplinaire en Cybersécurité (GRIC)(跨学科网络安全研究组(GRIC));Department of Computer Science, Université de Sherbrooke(谢布鲁克大学计算机科学系);Department of Management School, Université de Sherbrooke(谢布鲁克大学管理学院)
Massachusetts Institute of Technology(麻省理工学院);Stanford University(斯坦福大学);Google AI(谷歌人工智能)
我们调查是否事后模型的解释是有效的诊断模型错误-模型调试。为了应对解释模型预测的挑战,已经提出了大量的解释方法。尽管使用越来越多,但尚不清楚它们是否有效。首先,我们根据其来源将\textit{bugs}分类为:~ \textit{data, model, and test-time}污染缺陷。对于几种解释方法,我们评估它们的能力:检测虚假的相关性伪影(数据污染),诊断错误标记的训练样本(数据污染),区分(部分)重新初始化的模型和训练的模型(模型污染),以及检测分布外的输入(测试时污染)。我们发现,测试的方法能够诊断一个虚假的背景错误,但不能最终确定错误标记的训练样本。此外,一类修改反向传播算法的方法对于深度网络的高层参数是不变的;因此,对于诊断模型污染是无效的。我们补充我们的分析与人类受试者的研究,发现受试者无法识别有缺陷的模型使用属性,而是主要依赖于模型预测。总之,我们的研究结果提供了指导的从业者和研究人员转向解释模型调试的工具。
Peoples Water Data: Enabling Reliable Field Data Generation and Microbial Contamination Screening in Household Drinking Water
DST Unit of Nanoscience and Thematic Unit of Excellence, Department of Chemistry, IIT Madras(印度理工学院马德拉斯分校化学系DST纳米科学与卓越主题单元);School of Mechanical Engineering, Faculty of Engineering, Tel Aviv University(特拉维夫大学工程学院机械工程系);Faculty of Statistics, Tel Aviv University(特拉维夫大学统计学院);International Centre for Clean Water(国际清洁水中心)