# CounterBench: Evaluating and Improving Counterfactual Reasoning in Large Language Models

arXiv: https://arxiv.org/abs/2502.11008 · HTML: https://mlsanity.com/paper/2502.11008/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- Placed in Causal reasoning in LLMs, one of 17 directions in LLM reasoning.
- Builds on 15 of its 45 references that are in the map.
- First posted on arXiv 16 Feb 2025.

## Placed in

- LLM reasoning / [Causal reasoning in LLMs](https://mlsanity.com/lr/causal-reasoning-in-llms/)

## Builds on (most cited first, 15 of 15)

- [Sparks of Artificial General Intelligence: Early experiments with GPT-4](https://mlsanity.com/paper/2303.12712/) (2023)
- [Language Models as Knowledge Bases?](https://mlsanity.com/paper/1909.01066/) (2019)
- [Causal Reasoning and Large Language Models: Opening a New Frontier for Causality](https://mlsanity.com/paper/2305.00050/) (2023)
- [Causal Parrots: Large Language Models May Talk Causality But Are Not Causal](https://mlsanity.com/paper/2308.13067/) (2023)
- [Mapping the Increasing Use of LLMs in Scientific Papers](https://mlsanity.com/paper/2404.01268/) (2024)
- [DoWhy: An End-to-End Library for Causal Inference](https://mlsanity.com/paper/2011.04216/) (2020)
- [Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models](https://mlsanity.com/paper/2308.10379/) (2023)
- [CausalML: Python Package for Causal Machine Learning](https://mlsanity.com/paper/2002.11631/) (2020)
- [Understanding Causality with Large Language Models: Feasibility and Opportunities](https://mlsanity.com/paper/2304.05524/) (2023)
- [Causal Inference with Large Language Model: A Survey](https://mlsanity.com/paper/2409.09822/) (2024)
- [Large Language Models and Causal Inference in Collaboration: A Survey](https://mlsanity.com/paper/2403.09606/) (2024)
- [Counterfactual Token Generation in Large Language Models](https://mlsanity.com/paper/2409.17027/) (2024)
- [The potential of LLMs for coding with low-resource and domain-specific programming languages](https://mlsanity.com/paper/2307.13018/) (2023)
- [Causal Agent based on Large Language Model](https://mlsanity.com/paper/2408.06849/) (2024)
- [Relation-First Modeling Paradigm for Causal Representation Learning toward the Development of AGI](https://mlsanity.com/paper/2307.16387/) (2023)

## Built on by (most cited first, 21 of 21)

- [Thinking with Images for Multimodal Reasoning: Foundations, Methods, and Future Frontiers](https://mlsanity.com/paper/2506.23918/) (2025)
- [Simulating Society Requires Simulating Thought](https://mlsanity.com/paper/2506.06958/) (2025)
- [Executable Counterfactuals: Improving LLMs' Causal Reasoning Through Code](https://mlsanity.com/paper/2510.01539/) (2025)
- [MedGuideX: Internalizing Decision Logic from Executable Guidelines into Large Language Models for Clinical Reasoning](https://mlsanity.com/paper/2605.26567/) (2026)
- [Flying Pigs, FaR and Beyond: Evaluating LLM Reasoning in Counterfactual Worlds](https://mlsanity.com/paper/2505.22318/) (2025)
- [Distilling Counterfactual Reasoning from Language to Vision: Causal Graph Guided Post-Training for Video Understanding](https://mlsanity.com/paper/2511.19923/) (2025)
- [TempoBench: Reasoning Execution Without Causal Attribution Is Just Simulation](https://mlsanity.com/paper/2510.27544/) (2025)
- [Towards Generalizable Reasoning: Group Causal Counterfactual Policy Optimization for LLM Reasoning](https://mlsanity.com/paper/2602.06475/) (2026)
- [Evaluating Counterfactual Strategic Reasoning in Large Language Models](https://mlsanity.com/paper/2603.19167/) (2026)
- [ReplaySCM: A Benchmark for Executable Causal Mechanism Induction from Interventions](https://mlsanity.com/paper/2605.08197/) (2026)
- [CRiT-QA: Evaluating Multi-hop Reasoning with Counterfactual Chains and Distractor Traps](https://mlsanity.com/paper/2607.10562/) (2026)
- [From Language to Logic: A Bi-Level Framework for Structured Reasoning](https://mlsanity.com/paper/2507.08501/) (2025)
- [Better Think Thrice: Learning to Reason Causally with Double Counterfactual Consistency](https://mlsanity.com/paper/2602.16787/) (2026)
- [DeepImagine: Clinical Trial Outcome Prediction via Stepwise Local Counterfactual Imaginations](https://mlsanity.com/paper/2604.23054/) (2026)
- [Why We Need World Models for AGI: Where LLMs Fail and How World Models May Outperform](https://mlsanity.com/paper/2605.23972/) (2026)
- [Towards a Universal Causal Reasoner](https://mlsanity.com/paper/2605.24873/) (2026)
- [WorldKernel: A World Model is the Coupling Kernel of Admissible Possible Worlds](https://mlsanity.com/paper/2606.10934/) (2026)
- [The Illusion of textitWhat If: Evaluating the Breakdown of Counterfactual Reasoning in LLMs](https://mlsanity.com/paper/2608.27953/) (2026)
- [Everything in Moderation: Per-Domain Coverage Optima and Alignment-Resistant Domain Gaps in Multi-Domain Mid-Training](https://mlsanity.com/paper/2609.09081/) (2026)
- [Beyond Appearance Shifts: Task-Semantic Action Calibration for VLA Models](https://mlsanity.com/paper/2609.23650/) (2026)
- [CoBe: A Benchmark For Conversational Counterfactual Text Editing](https://mlsanity.com/paper/2610.13057/) (2026)

## Abstract (arXiv)

Counterfactual reasoning is widely recognized as one of the most challenging and intricate aspects of causality in artificial intelligence. In this paper, we evaluate the performance of large language models (LLMs) in counterfactual reasoning. In contrast to previous studies that primarily focus on commonsense causal reasoning, where LLMs often rely on prior knowledge for inference, we specifically assess their ability to perform counterfactual inference using a set of formal rules. To support this evaluation, we introduce a new benchmark dataset, CounterBench, comprising 1K counterfactual reasoning questions. The dataset is designed with varying levels of difficulty, diverse causal graph structures, distinct types of counterfactual questions, and multiple nonsensical name variants. Our experiments demonstrate that counterfactual reasoning poses a significant challenge for LLMs, with most models performing at levels comparable to random guessing. To enhance LLM's counterfactual reasoning ability, we propose a novel reasoning paradigm, CoIn, which guides LLMs through iterative reasoning and backtracking to systematically explore counterfactual solutions. Experimental results show that our method significantly improves LLM performance on counterfactual reasoning tasks and consistently enhances performance across different LLMs.Our dataset is available at https://huggingface.co/datasets/CounterBench/CounterBench.