Large language models (LLMs) increasingly act as data-science agents, combining abstract reasoning with advanced tool use. Yet the relevant benchmark landscape largely divides into symbolic causal reasoning benchmarks without realistic data analysis or data analysis benchmarks without causal data-generating structure or a range of causal data science tasks. Existing causal evaluation datasets are often restricted to curated examples from existing sources, with diversity coming from limited templatized variations rather than from systematic generation of novel synthetic causal structures. We introduce CausalDS, a benchmark for evaluating causal reasoning in agentic data-science workflows. Each benchmark instance is a scene consisting of a sampled structural causal model (SCM) with generated observational data and an accompanying synthetic natural-language story grounded in a real domain. We ground the composition of the benchmark components in empirical distributions obtained from real-world datasets, thus retaining empirical structure while reducing the memorization risk through completely synthetic generation. From each scene, we then derive tasks spanning all three of Pearl's rungs. Most tasks include a data science coding component, where the model typically needs to use several tools to arrive at the final answer due to imperfect observations. Abstention is tested directly: for targets the graph does not identify, the graded answer is to abstain. The benchmark thus jointly evaluates symbolic causal reasoning, data science, uncertainty quantification, abstention, and tool use/coding. Because the generator is fully synthetic, the exam is not a fixed dataset: composition, difficulty, and size are parameters, and fresh exams can be drawn at will.
There are many different causal effect estimators in causal inference. However, it is unclear how to choose between these estimators because there is no ground-truth for causal effects. A commonly used option is to simulate synthetic data, where the ground-truth is known. However, the best causal estimators on synthetic data are unlikely to be the best causal estimators on real data. An ideal benchmark for causal estimators would both (a) yield ground-truth values of the causal effects and (b) be representative of real data. Using flexible generative models, we provide a benchmark that both yi…
Causal relationships are commonly examined in manufacturing processes to support faults investigations, perform interventions, and make strategic decisions. Industry 4.0 has made available an increasing amount of data that enable data-driven Causal Discovery (CD). Considering the growing number of recently proposed CD methods, it is necessary to introduce strict benchmarking procedures on publicly available datasets since they represent the foundation for a fair comparison and validation of different methods. This work introduces two novel public datasets for CD in continuous manufacturing pro…
CausalGame: Benchmarking Causal Thinking of LLM Agents in Games
MBZUAI（穆罕默德·本·扎耶德人工智能大学）；Carnegie Mellon University（卡内基梅隆大学）；Hong Kong Baptist University（香港浸会大学）；University of Oxford（牛津大学）；New York University, Abu Dhabi（纽约大学阿布扎比分校）
Causal inference analysis is the estimation of the effects of actions on outcomes. In the context of healthcare data this means estimating the outcome of counter-factual treatments (i.e. including treatments that were not observed) on a patient's outcome. Compared to classic machine learning methods, evaluation and validation of causal inference analysis is more challenging because ground truth data of counter-factual outcome can never be obtained in any real-world scenario. Here, we present a comprehensive framework for benchmarking algorithms that estimate causal effect. The framework includ…
In this paper we present a comprehensive view of prominent causal discovery algorithms, categorized into two main categories (1) assuming acyclic and no latent variables, and (2) allowing both cycles and latent variables, along with experimental results comparing them from three perspectives: (a) structural accuracy, (b) standard predictive accuracy, and (c) accuracy of counterfactual inference. For (b) and (c) we train causal Bayesian networks with structures as predicted by each causal discovery technique to carry out counterfactual or standard predictive inference. We compare causal algorit…
Massachusetts Institute of Technology（麻省理工学院）；California Institute of Technology（加州理工学院）；Amazon Web Services（亚马逊云科技）
Measuring algorithmic bias is crucial both to assess algorithmic fairness, and to guide the improvement of algorithms. Current methods to measure algorithmic bias in computer vision, which are based on observational datasets, are inadequate for this task because they conflate algorithmic bias with dataset bias. To address this problem we develop an experimental method for measuring algorithmic bias of face analysis algorithms, which manipulates directly the attributes of interest, e.g., gender and skin tone, in order to reveal causal links between attribute variation and performance change. Ou…
AutoMatBench: An Automatic Optimization Toolkit for the Acceleration of Material Properties Prediction Benchmarking
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China（中国科学院计算技术研究所）；University of Chinese Academy of Sciences, Beijing, China（中国科学院大学）
Towards Real-world Human Behavior Simulation: Benchmarking Large Language Models on Long-horizon, Cross-scenario, Heterogeneous Behavior Traces
Chinese Information Processing Laboratory, Institute of Software, Chinese Academy of Sciences（中国科学院软件研究所信息处理实验室）；University of Chinese Academy of Sciences（中国科学院大学）；Kuaishou Technology（快手科技）
City University of Hong Kong（香港城市大学）；Arizona State University（亚利桑那州立大学）
The convenient access to copious multi-faceted data has encouraged machine learning researchers to reconsider correlation-based learning and embrace the opportunity of causality-based learning, i.e., causal machine learning (causal learning). Recent years have therefore witnessed great effort in developing causal learning algorithms aiming to help AI achieve human-level intelligence. Due to the lack-of ground-truth data, one of the biggest challenges in current causal learning research is algorithm evaluations. This largely impedes the cross-pollination of AI and causal inference, and hinders …
Simulators make unique benchmarks for causal effect estimation since they do not rely on unverifiable assumptions or the ability to intervene on real-world systems, but are often too simple to capture important aspects of real applications. We propose a simulator of Alzheimer's disease aimed at modeling intricacies of healthcare data while enabling benchmarking of causal effect and policy estimators. We fit the system to the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset and ground hand-crafted components in results from comparative treatment trials and observational treatment patt…
MedSNIP: Building and Benchmarking Snippet-Level Granularity for Medical Fact Verification
Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）；Yale University（耶鲁大学）；Korea University（高丽大学）；Beijing Normal University（北京师范大学）；INSAIT, Sofia University, “St. Kliment Ohridski”（索非亚大学INSAIT研究所）
Benchmarking Dependence Measures to Prevent Shortcut Learning in Medical Imaging
Hertie Institute for AI in Brain Health, University of Tübingen（蒂宾根大学赫蒂脑健康人工智能研究所）；University Hospital of Tübingen（蒂宾根大学医院）；University of Stuttgart（斯图加特大学）；University of Bern（伯尔尼大学）；Inselspital, Bern University Hospital（伯尔尼大学医院 Inselspital）；Stanford University（斯坦福大学）
GT-HarmBench: Benchmarking AI Safety Risks Through the Lens of Game Theory
ETH Zürich（苏黎世联邦理工学院）；Berea College（贝雷学院）；University of Toronto（多伦多大学）；Vector Institute（向量研究所）；Max Planck Institute for Intelligent Systems, Tübingen, Germany（图宾根德国智能系统马克斯·普朗克研究所）
In this paper, we propose a gamification approach as a novel framework for smart building infrastructure with the goal of motivating human occupants to reconsider personal energy usage and to have positive effects on their environment. Human interaction in the context of cyber-physical systems is a core component and consideration in the implementation of any smart building technology. Research has shown that the adoption of human-centric building services and amenities leads to improvements in the operational efficiency of these cyber-physical systems directed towards controlling building ene…
GraIP: A Benchmarking Framework For Neural Graph Inverse Problems
Mila -- Quebec AI Institute, Montréal, Canada（魁北克AI研究所）；Department of Computer Science, University of Stuttgart, Germany（斯图加特大学计算机科学系）；International Max Planck Research School for Intelligent Systems, German（智能系统国际马克斯·普朗克研究学校）；ETAS Research, Germany（ETAS研究）；Faculty of Computer Science, RWTH Aachen University, Germany（亚琛工业大学计算机科学系）；NEC Laboratories Europe（NEC欧洲实验室）
InEdit-Bench: Benchmarking Intermediate Logical Pathways for Intelligent Image Editing Models
State Key Laboratory of Optical Field Manipulation Science and Technology, Institute of Optics and Electronics, Chinese Academy of Sciences（光学场操控科学与技术国家重点实验室，光学电子研究所，中国科学院）；National Laboratory on Adaptive Optics（自适应光学国家实验室）；University of Chinese Academy of Sciences（中国科学院大学）
University of California, Berkeley（加州大学伯克利分校）；Harvard Society of Fellows（哈佛大学学者协会）；Max-Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）
Dynamic benchmarks interweave model fitting and data collection in an attempt to mitigate the limitations of static benchmarks. In contrast to an extensive theoretical and empirical study of the static setting, the dynamic counterpart lags behind due to limited empirical studies and no apparent theoretical foundation to date. Responding to this deficit, we initiate a theoretical study of dynamic benchmarking. We examine two realizations, one capturing current practice and the other modeling more complex settings. In the first model, where data collection and model fitting alternate sequentiall…
