Recent work suggests that large language models may implicitly learn world models. How should we assess this possibility? We formalize this question for the case where the underlying reality is governed by a deterministic finite automaton. This includes problems as diverse as simple logical reasoning, geographic navigation, game-playing, and chemistry. We propose new evaluation metrics for world model recovery inspired by the classic Myhill-Nerode theorem from language theory. We illustrate their utility in three domains: game playing, logic puzzles, and navigation. In all domains, the generative models we consider do well on existing diagnostics for assessing world models, but our evaluation metrics reveal their world models to be far less coherent than they appear. Such incoherence creates fragility: using a generative model to solve related but subtly different tasks can lead to failures. Building generative models that meaningfully capture the underlying logic of the domains they model would be immensely valuable; our results suggest new ways to assess how close a given model is to that goal.
A word association network methodology for evaluating implicit biases in LLMs compared to humans
University of Pisa, Department of Computer Science(比萨大学计算机科学系);National Research Council of Italy, Institute of Information Science and Technologies(意大利国家研究委员会信息科学与技术研究所);University of Trento, Department of Psychology and Cognitive Science(特伦托大学心理学与认知科学系)
Evaluating Implicit Bias in Large Language Models by Attacking From a Psychometric Perspective
State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences(中国科学院人工智能安全国家重点实验室,计算技术研究所);Key Laboratory of Network Data Science and Technology, Institute of Computing Technology, Chinese Academy of Sciences(中国科学院网络数据科学与技术重点实验室,计算技术研究所);University of Chinese Academy of Sciences(中国科学院大学)
Scene Graph-Guided Generative AI Framework for Synthesizing and Evaluating Industrial Hazard Scenarios
Ph.D. Student, Dept. of Civil Eng., University of Texas at Arlington. E-mail;Assistant Professor, Dept. of Computer Sci. \& Eng., University of Texas at Arlington. E-mail;Assistant Professor, Dept. of Civil Eng., University of Texas at Arlington. E-mail
MDIT-Bench: Evaluating the Dual-Implicit Toxicity in Large Multimodal Models
Harbin Institute of Technology (Shenzhen)(哈尔滨工业大学(深圳));Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies(广东省新型安全智能技术重点实验室);University of Barcelona(巴塞罗那大学)
Explicit vs. Implicit Biographies: Evaluating and Adapting LLM Information Extraction on Wikidata-Derived Texts
DISI Departement of Computer Science and Engineering, University of Bologna(博洛尼亚大学计算机科学与工程系);Automobili Lamborghini SpA(法拉利公司);Digital Humanities Advanced Research Center(数字人文高级研究中心);KNAW Humanities Cluster, DHLab(KNAW 人文集群,DHLab);ETH Zurich, Collegium Helveticum(苏黎世联邦理工学院,瑞士学院)
Evaluating Generative Ad Hoc Information Retrieval
Leipzig University(莱比锡大学);The University of Queensland(昆士兰大学);Bauhaus-Universität Weimar(魏玛包豪斯大学);Friedrich-Schiller-Universität Jena(弗里德里希·席勒大学耶拿分校);University of Kassel(卡塞尔大学)