AniDriveQA: a VQA dataset for driving scenes with animal presence
The experimental results demonstrate that AniDriveQA effectively exposes the limitations of current VLMs in rare yet safety-critical autonomous driving scenarios. The dataset provides a valuable diagnostic benchmark for advancing reasoning, perception, and decision-making capabilities in future vision-language models.
Authors
- Huai Yu (ORCID: https://orcid.org/0000-0001-6043-3412)
- Ruiqi Wang (ORCID: https://orcid.org/0000-0003-0424-6799)
- Hao Hu (ORCID: https://orcid.org/0000-0002-8859-0986)
- Rui Wang (ORCID: https://orcid.org/0009-0000-6178-9808)
Institutions
- China Academy of Railway Sciences (CN)
- Xi'an Jiaotong University (CN)
Publication Details
- Journal
- Frontiers in Robotics and AI
- Published
- 2025-10-28
- DOI
- https://doi.org/10.3389/frobt.2025.1684845
- Primary Topic
- Multimodal Machine Learning Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- China Railway
- National Natural Science Foundation of China
- China Academy of Railway Sciences