CausalGame: Benchmarking Causal Thinking of LLM Agents in Games

• URL: http://arxiv.org/abs/2607.04293v1
• Date: Sun, 05 Jul 2026 13:08:51 GMT
• Title: CausalGame: Benchmarking Causal Thinking of LLM Agents in Games
• Abstract summary: CausalGame is a benchmark for evaluating the causal thinking capabilities of AI Scientists.<n>We design 14 scenarios that incorporate selection bias, measurement error, and hidden confounders.<n>The best model reaches only 68.0% survival against analytical optima of 78-85%, and merely 5-7% of sessions receive credits on the causal-reasoning rubrics.
• Score: 55.189910630332065
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Building AI Scientist agents with Large Language Models (LLMs) has recently attracted growing attention. Since scientific discovery fundamentally relies on uncovering causal relationships from observations, the capability of causal thinking, i.e., distinguishing causation from correlation and recognizing hidden biases, is essential to LLM agents. Although a number of benchmarks exist for AI Scientists, none explicitly incorporate challenges from selection bias, measurement error, and hidden confounders that widely exist in real-world scientific discovery. To this end, we present CausalGame, a benchmark that evaluates the causal thinking capabilities of LLM agents through interactive games. CausalGame asks LLM agents to actively design experimental protocols, collect observation data, and derive a final solution with an explanation report. To emulate realistic scientific discovery challenges, we design 14 scenarios that incorporate selection bias, measurement error, and hidden confounders. Across 30 LLM agents, none demonstrates reliable causal thinking: the best model reaches only 68.0% survival against analytical optima of 78-85%, and merely 5-7% of sessions receive credits on the causal-reasoning rubrics. CausalGame provides a scalable and controlled testbed for evaluating the causal thinking of AI Scientist agents.

Related papers

• (Human) Attention Is (Still) All You Need: Human oversight makes AI-assisted social science reliable [15.442870614597451]
  We argue that the reliability of AI-assisted research depends not only on model capability, but also on how cognitive labour is structured between humans and machines.<n>We study this problem through Human-in-the-Loop Economic Research (HLER), a decision architecture based on pre-commitment, decision sequencing, accountability, and attention allocation.
  arXiv  Detail & Related papers  (2026-06-11T03:22:24Z)
• Causal Ensemble Agent: Hierarchical Causal Discovery with LLM-guided Expert Reweighting [95.24017293537025]
  Causal Ensemble Agent (CEA) is a novel framework that aggregates structural insights from statistical discovery experts across different graph levels.<n>CEA achieves the strongest overall performance across a wide range of causal discovery methods.
  arXiv  Detail & Related papers  (2026-06-09T09:09:07Z)
• CausaLab: A Scalable Environment for Interactive Causal Discovery Toward AI Scientists [28.253879252786632]
  We introduce CausaLab, a scalable environment for evaluating interactive causal discovery by LLM agents.<n>Unlike prior evaluations, CausaLab evaluates both whether an agent can solve a problem using causal evidence and whether its answer is grounded in a faithful recovered causal mechanism.
  arXiv  Detail & Related papers  (2026-05-25T16:57:06Z)
• FIRE-Bench: Evaluating Agents on the Rediscovery of Scientific Insights [63.32178443510396]
  We introduce FIRE-Bench (Full-cycle Insight Rediscovery Evaluation), a benchmark that evaluates agents through the rediscovery of established findings.<n>Even the strongest agents achieve limited rediscovery success (50 F1), exhibit high variance across runs, and display recurring failure modes in experimental design, execution, and evidence-based reasoning.
  arXiv  Detail & Related papers  (2026-02-02T23:21:13Z)
• ResearchBench: Benchmarking LLMs in Scientific Discovery via Inspiration-Based Task Decomposition [67.26124739345332]
  Large language models (LLMs) have demonstrated potential in assisting scientific research, yet their ability to discover high-quality research hypotheses remains unexamined.<n>We introduce the first large-scale benchmark for evaluating LLMs with a near-sufficient set of sub-tasks of scientific discovery.<n>We develop an automated framework that extracts critical components - research questions, background surveys, inspirations, and hypotheses - from scientific papers.
  arXiv  Detail & Related papers  (2025-03-27T08:09:15Z)
• Can Large Language Models Help Experimental Design for Causal Discovery? [94.66802142727883]
  Large Language Model Guided Intervention Targeting (LeGIT) is a robust framework that effectively incorporates LLMs to augment existing numerical approaches for the intervention targeting in causal discovery.<n>LeGIT demonstrates significant improvements and robustness over existing methods and even surpasses humans.
  arXiv  Detail & Related papers  (2025-03-03T03:43:05Z)
• Auto-Bench: An Automated Benchmark for Scientific Discovery in LLMs [23.608962459019278]
  We introduce a novel benchmark to evaluate Large Language Models (LLMs) for scientific discovery in both natural and social sciences.<n>Our benchmark is based on the principles of causal graph discovery. It challenges models to uncover hidden structures and make optimal decisions, which includes generating valid justifications.<n>We evaluate state-of-the-art LLMs, including GPT-4, Gemini, Qwen, Claude, and Llama, and observe a significant performance drop as the problem complexity increases.
  arXiv  Detail & Related papers  (2025-02-21T05:35:20Z)
• COLD: Causal reasOning in cLosed Daily activities [7.782872276680731]
  We propose the COLD (Causal reasOning in cLosed Daily activities) framework.<n>It is built upon human understanding of daily real-world activities to reason about the causal nature of events.<n>We show that the proposed framework facilitates the creation of enormous causal queries.
  arXiv  Detail & Related papers  (2024-11-29T06:37:13Z)
• Context-Aware Reasoning On Parametric Knowledge for Inferring Causal Variables [49.31233968546582]
  We introduce a novel benchmark where the objective is to complete a partial causal graph.<n>We show the strong ability of LLMs to hypothesize the backdoor variables between a cause and its effect.<n>Unlike simple memorization of fixed associations, our task requires the LLM to reason according to the context of the entire graph.
  arXiv  Detail & Related papers  (2024-09-04T10:37:44Z)
• On the Reliability of Large Language Models for Causal Discovery [55.94868919310357]
  This study investigates the efficacy of Large Language Models (LLMs) in causal discovery.<n>We use newly available open-source LLMs, OLMo and BLOOM, which provide access to their pre-training corpora.<n>We examine: (i) the impact of memorization for accurate causal relation prediction, (ii) the influence of incorrect causal relations in pre-training data, and (iii) the contextual nuances that influence LLMs' understanding of causal relations.
  arXiv  Detail & Related papers  (2024-07-29T01:45:05Z)
• Discovering and Reasoning of Causality in the Hidden World with Large Language Models [109.62442253177376]
  We develop a new framework termed Causal representatiOn AssistanT (COAT) to propose useful measured variables for causal discovery.<n>Instead of directly inferring causality with Large language models (LLMs), COAT constructs feedback from intermediate causal discovery results to LLMs to refine the proposed variables.
  arXiv  Detail & Related papers  (2024-02-06T12:18:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.