Website Repository of "Retrieval-Based Interleaved Visual Chain-of-Thought in Real-World Driving Scenarios"

• The DrivingVQA dataset is available on HuggingFace Hub.
• The code of RIV-CoT is available at https://github.com/vita-epfl/RIV-CoT.

Citation

If you use the DrivingVQA dataset or RIV-CoT in your research, please cite our paper:

@misc{drivingvqa2025,
      title         = {Retrieval-Based Interleaved Visual Chain-of-Thought in Real-World Driving Scenarios},
      author        = {Corbière, Charles and Roburin, Simon and Montariol, Syrielle and Bosselut, Antoine and Alahi, Alexandre},
      year          = {2025},
      eprint        = {2501.04671},
      archivePrefix = {arXiv},
      primaryClass  = {cs.CV},
}