Papers/2609.01659
🧪 Test?View on arXiv

Beyond Textual Chain-of-Thought: A Survey on Action-Grounded Reasoning in Autonomous Driving

Tang Zhengxu, Author 2, Author 3, Author 4, Author 5

action-grounded reasoningautonomous drivingintermediate representationssafety-critical systems
2609.01659
Builder Relevance
80%
2h ago

Abstract

This survey explores the transition from textual chain-of-thought reasoning to action-grounded reasoning in autonomous driving.

Reality Card

Core Claim

The paper proposes a representation-centered taxonomy for action-grounded reasoning, categorizing 130 methods into four main categories and identifying the need for intermediate representations that are grounded in the real world.

Method / Result

Surveyed 171 papers, including 130 method papers and 41 benchmarks.

Limitations

The main limitation is the challenge of grounding intermediate representations in real-world scenarios while ensuring safety in action execution.

Paper to code

Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.

No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.
← Back to all papers