# Causal reasoning in LLMs

Area: [LLM reasoning](https://mlsanity.com/lr/index.md). HTML: https://mlsanity.com/lr/causal-reasoning-in-llms/

## Key facts

- Causal reasoning in LLMs: 167 papers on arXiv since 2022 (data through 14 Oct 2026).
- 40 of its 167 mapped papers (24%) cite “CLadder: Assessing Causal Reasoning in Language Models” (2023, arXiv 2312.04350).
- 37 of its 167 mapped papers (22%) cite “Causal Reasoning and Large Language Models: Opening a New Frontier for Causality” (2023, arXiv 2305.00050).
- 5 new papers 17 Sep–14 Oct.

## Most cited here

- [CLadder: Assessing Causal Reasoning in Language Models](https://mlsanity.com/paper/2312.04350/) (2023), cited by 40 papers here
- [Causal Reasoning and Large Language Models: Opening a New Frontier for Causality](https://mlsanity.com/paper/2305.00050/) (2023), cited by 37 papers here
- [e-CARE: a New Dataset for Exploring Explainable Causal Reasoning](https://mlsanity.com/paper/2205.05849/) (2022), cited by 13 papers here
- [Unveiling Causal Reasoning in Large Language Models: Reality or Mirage?](https://mlsanity.com/paper/2506.21215/) (2025), cited by 13 papers here
- [Is ChatGPT a Good Causal Reasoner? A Comprehensive Evaluation](https://mlsanity.com/paper/2305.07375/) (2023), cited by 10 papers here
- [Are LLMs Capable of Data-based Statistical and Causal Reasoning? Benchmarking Advanced Quantitative Reasoning with Data](https://mlsanity.com/paper/2402.17644/) (2024), cited by 10 papers here
- [Causal Evaluation of Language Models](https://mlsanity.com/paper/2405.00622/) (2024), cited by 9 papers here
- [CLEAR: Can Language Models Really Understand Causal Graphs?](https://mlsanity.com/paper/2406.16605/) (2024), cited by 7 papers here
- [CounterBench: Evaluating and Improving Counterfactual Reasoning in Large Language Models](https://mlsanity.com/paper/2502.11008/) (2025), cited by 7 papers here
- [COLA: Contextualized Commonsense Causal Reasoning from the Causal Inference Perspective](https://mlsanity.com/paper/2305.05191/) (2023), cited by 6 papers here

## New (17 Sep–14 Oct)

- [Code-Interactive Visual Reasoning for Financial Candlestick Understanding](https://mlsanity.com/paper/2610.16587/) (2026-10-12)
- [Reasoning Models Are Accurate but Unsound on Identification](https://mlsanity.com/paper/2610.03519/) (2026-10-02)
- [Thinking Leakage: A Causal Audit of NoThink Post-Training in Hybrid Reasoning Models](https://mlsanity.com/paper/2609.28682/) (2026-09-23)
- [Are Stated Reasoning Steps Causally Load-Bearing?](https://mlsanity.com/paper/2609.27038/) (2026-09-22)
- [xWhyL: Causal Interactive Learning](https://mlsanity.com/paper/2609.26037/) (2026-09-22)