# CausalDriveBench: Evaluating Causal Reasoning in Vision-Language-Action Models for Autonomous Driving

arXiv: https://arxiv.org/abs/2609.32157 · HTML: https://mlsanity.com/paper/2609.32157/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Chain-of-thought reasoning for driving, one of 17 directions in LLM reasoning.
- Builds on 16 of its 45 references that are in the map.
- First posted on arXiv 26 Sep 2026.

## Placed in

- LLM reasoning / [Chain-of-thought reasoning for driving](https://mlsanity.com/lr/chain-of-thought-reasoning-for-driving/)
- Vision-language-action models / [VLA for autonomous driving](https://mlsanity.com/vla/vla-for-autonomous-driving/)
- End-to-end autonomous driving / [Reasoning VLA and VLM driving policies](https://mlsanity.com/driving/reasoning-vla-and-vlm-driving-policies/)

## Builds on (most cited first, 16 of 16)

- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecasting](https://mlsanity.com/paper/2301.00493/) (2023)
- [EMMA: End-to-End Multimodal Model for Autonomous Driving](https://mlsanity.com/paper/2410.23262/) (2024)
- [ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Driving](https://mlsanity.com/paper/2506.08052/) (2025)
- [Cosmos-Reason1: From Physical Common Sense To Embodied Reasoning](https://mlsanity.com/paper/2503.15558/) (2025)
- [Alpamayo-R1: Bridging Reasoning and Action Prediction for Generalizable Autonomous Driving in the Long Tail](https://mlsanity.com/paper/2511.00088/) (2025)
- [DriveMoE: Mixture-of-Experts for Vision-Language-Action Model in End-to-End Autonomous Driving](https://mlsanity.com/paper/2505.16278/) (2025)
- [A Survey on Vision-Language-Action Models for Autonomous Driving](https://mlsanity.com/paper/2506.24044/) (2025)
- [Impromptu VLA: Open Weights and Open Data for Driving Vision-Language-Action Models](https://mlsanity.com/paper/2505.23757/) (2025)
- [WiseAD: Knowledge Augmented End-to-End Autonomous Driving with Vision-Language Model](https://mlsanity.com/paper/2412.09951/) (2024)
- [SafeAuto: Knowledge-Enhanced Safe Autonomous Driving with Multimodal Foundation Models](https://mlsanity.com/paper/2503.00211/) (2025)
- [Counterfactual VLA: Self-Reflective Vision-Language-Action Model with Adaptive Reasoning](https://mlsanity.com/paper/2512.24426/) (2025)
- [UniDriveVLA: Unifying Understanding, Perception, and Action Planning for Autonomous Driving](https://mlsanity.com/paper/2604.02190/) (2026)
- [DriveAction: A Benchmark for Exploring Human-like Driving Decisions in VLA Models](https://mlsanity.com/paper/2506.05667/) (2025)
- [OpenREAD: Reinforced Open-Ended Reasoning for End-to-End Autonomous Driving with LLM-as-Critic](https://mlsanity.com/paper/2512.01830/) (2025)
- [CoC-VLA: Delving into Adversarial Domain Transfer for Explainable Autonomous Driving via Chain-of-Causality Visual-Language-Action Model](https://mlsanity.com/paper/2511.19914/) (2025)

## Abstract (arXiv)

Vision-Language-Action (VLA) models for autonomous driving produce natural-language reasoning alongside predicted trajectories, but whether this reasoning reflects the causal structure of the scene remains untested. We introduce CausalDriveBench, an evaluation framework grounded in Pearl's Causal Hierarchy (PCH) that tests causal reasoning in driving-specific VLAs through structured visual question answering (QA) and alternative-trajectory prediction. To this end, we construct causal scene graphs over nuScenes that distinguish causally active, dormant, and distractor entities, separating perceptual salience from causal relevance. The benchmark spans all four rungs of PCH (association, intervention, and counterfactual along with causal discovery) for QA generation. For the higher rungs, we additionally provide reference trajectories under specified scene modifications, enabling action-level verification that complements reasoning-level evaluation. In total, the benchmark contains 7,285 verified causal QA pairs and 1,000 counterfactual trajectories derived from nuScenes. We evaluate 10 driving-specific VLAs and 3 general-purpose VLMs, and report three findings. First, the best model reaches only 70.6% QA accuracy, and 4 of 13 models score below random chance. Second, comparing each driving VLA to the general-purpose VLM that shares its language backbone, the cost of driving fine-tuning ranges from 2 to 34 percentage points on causal QA, with post-training design explaining the spread. Third, causal QA and trajectory accuracy are statistically uncorrelated across models: under counterfactual prompts, predicted trajectories either over-react or collapse onto the observed-scene baseline. Taken together, these results show that neither fluent rationales nor accurate observed-scene trajectories constitute evidence of causal understanding.