Do Large Language Models Show Biases in Causal Learning?
UBA（布宜诺斯艾利斯大学）；Università degli Studi di Genova（热那亚大学）；University of Buenos Aires（布宜诺斯艾利斯大学）；Universidad Nacional del Sur（南方国立大学）；IIIA-CSIC（西班牙国家研究委员会人工智能研究所）；Universidad Autónoma de Barcelona（巴塞罗那自治大学）；Arizona State University（亚利桑那州立大学）
Maria Victoria Carro, Francisca Gauna Selasco, Denise Alejandra Mester, Margarita Gonzales, Mario A. Leiva, Maria Vanina Martinez, Gerardo I. Simari
Causal learning is the cognitive process of developing the capability of making causal inferences based on available information, often guided by normative principles. This process is prone to errors and biases, such as the illusion of causality, in which people perceive a causal relationship between two variables despite lacking supporting evidence. This cognitive bias has been proposed to underlie many societal problems, including social prejudice, stereotype formation, misinformation, and superstitious thinking. In this research, we investigate whether large language models (LLMs) develop causal illusions, both in real-world and controlled laboratory contexts of causal learning and inference. To this end, we built a dataset of over 2K samples including purely correlational cases, situations with null contingency, and cases where temporal information excludes the possibility of causality by placing the potential effect before the cause. We then prompted the models to make statements or answer causal questions to evaluate their tendencies to infer causation erroneously in these structured settings. Our findings show a strong presence of causal illusion bias in LLMs. Specifically, in open-ended generation tasks involving spurious correlations, the models displayed bias at levels comparable to, or even lower than, those observed in similar studies on human subjects. However, when faced with null-contingency scenarios or temporal cues that negate causal relationships, where it was required to respond on a 0-100 scale, the models exhibited significantly higher bias. These findings suggest that the models have not uniformly, consistently, or reliably internalized the normative principles essential for accurate causal learning.
Prompting Fairness: Integrating Causality to Debias Large Language Models
Google DeepMind（谷歌DeepMind）；Carnegie Mellon University（卡内基梅隆大学）；University of Maryland, College Park（马里兰大学帕克分校）；Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）；University of Texas at Austin（德克萨斯大学奥斯汀分校）；University of California, Santa Cruz（加州大学圣克鲁兹分校）
Information Gain-Guided Causal Intervention for Autonomous Debiasing Large Language Models
Research Center for Social Computing and Interactive Robotics（社会计算与交互机器人研究中心）；Harbin Institute of Technology（哈尔滨工业大学）；Beijing Academy of Artificial Intelligence（北京人工智能研究院）
Quantifying and Mitigating Unimodal Biases in Multimodal Large Language Models: A Causal Perspective
Shanghai AI Laboratory（上海人工智能实验室）；State Key Laboratory of General Artificial Intelligence, Peking University（北京大学通用人工智能国家重点实验室）；School of Intelligence Science and Technology, Peking University（北京大学智能科学与技术学院）；School of Computer Science, Fudan University（复旦大学计算机科学技术学院）；Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）
CLadder: Assessing Causal Reasoning in Language Models
MPI for Intelligent Systems, Tübingen（马克斯·普朗克智能系统研究所，蒂宾根）；ETH Zürich（苏黎世联邦理工学院）；IIT Kharagpur（印度理工学院卡拉格普尔分校）；University of Hong Kong（香港大学）；University of Washington（华盛顿大学）
Causal Parrots: Large Language Models May Talk Causality But Are Not Causal
TU Darmstadt（达姆施塔特工业大学）；Hessian Center for AI (hessian.AI)（黑森人工智能中心）；Centre for Cognitive Science, TU Darmstadt（达姆施塔特工业大学认知科学中心）；German Research Center for Artificial Intelligence (DFKI)（德国人工智能研究中心）
Evaluation of biases in language models is often limited to synthetically generated datasets. This dependence traces back to the need for a prompt-style dataset to trigger specific behaviors of language models. In this paper, we address this gap by creating a prompt dataset with respect to occupations collected from real-world natural sentences present in Wikipedia. We aim to understand the differences between using template-based prompts and natural sentence prompts when studying gender-occupation biases in language models. We find bias evaluations are very sensitive to the design choices of …
To what extent do pre-trained language models grasp semantic knowledge regarding the phenomenon of distributivity? In this paper, we introduce DistNLI, a new diagnostic dataset for natural language inference that targets the semantic difference arising from distributivity, and employ the causal mediation analysis framework to quantify the model behavior and explore the underlying mechanism in this semantically-related task. We find that the extent of models' understanding is associated with model size and vocabulary size. We also provide insights into how models encode such high-level semantic…
University of Chinese Academy of Sciences（中国科学院大学）；Institute of Software, Chinese Academy of Sciences（中国科学院软件研究所）；Beijing Academy of Artificial Intelligence（北京智源人工智能研究院）
Prompt-based probing has been widely used in evaluating the abilities of pretrained language models (PLMs). Unfortunately, recent studies have discovered such an evaluation may be inaccurate, inconsistent and unreliable. Furthermore, the lack of understanding its inner workings, combined with its wide applicability, has the potential to lead to unforeseen risks for evaluating and applying PLMs in real-world applications. To discover, understand and quantify the risks, this paper investigates the prompt-based probing from a causal view, highlights three critical biases which could induce biased…
KU Leuven（鲁汶大学）；University of Antwerp（安特卫普大学）；TU Berlin（柏林工业大学）
An increasing awareness of biased patterns in natural language processing resources, like BERT, has motivated many metrics to quantify `bias' and `fairness'. But comparing the results of different metrics and the works that evaluate with such metrics remains difficult, if not outright impossible. We survey the existing literature on fairness metrics for pretrained language models and experimentally evaluate compatibility, including both biases in language models as in their downstream tasks. We do this by a mixture of traditional literature survey and correlation analysis, as well as by runnin…
Neuroimaging datasets keep growing in size to address increasingly complex medical questions. However, even the largest datasets today alone are too small for training complex machine learning models. A potential solution is to increase sample size by pooling scans from several datasets. In this work, we combine 12,207 MRI scans from 15 studies and show that simple pooling is often ill-advised due to introducing various types of biases in the training data. First, we systematically define these biases. Second, we detect bias by experimentally showing that scans can be correctly assigned to the…
Language Agents Meet Causality -- Bridging LLMs and Causal World Models
University of Amsterdam（阿姆斯特丹大学）；Institute for Logic, Language and Computation (ILLC)（逻辑、语言与计算研究所）；University of Edinburgh（爱丁堡大学）；Institute for Language, Cognition and Computation (ILCC)（语言、认知与计算研究所）
University of Southern California（南加州大学）；Carnegie Mellon University（卡内基梅隆大学）
Numerous works have analyzed biases in vision and pre-trained language models individually - however, less attention has been paid to how these biases interact in multimodal settings. This work extends text-based bias analysis methods to investigate multimodal language models, and analyzes intra- and inter-modality associations and biases learned by these models. Specifically, we demonstrate that VL-BERT (Su et al., 2020) exhibits gender biases, often preferring to reinforce a stereotype over faithfully describing the visual scene. We demonstrate these findings on a controlled case-study and e…
