Accepted paper ·
DriveDNA and BATON accepted to ACM KDD 2027
Two of the lab's naturalistic-driving benchmarks — DriveDNA for driving-style identification and BATON for human–automation control transitions — will be presented at ACM KDD 2027 in San Jose, California (August 2027).
Two papers from the lab, both first-authored by Yuhang Wang, have been accepted to the 33rd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2027), to be held in San Jose, California, on August 1–5, 2027:
- DriveDNA: A Large-Scale Multimodal Naturalistic Driving Dataset and Benchmark for Driving Style Identification (Yuhang Wang, Lingyao Li, Hao Zhou). The release covers 4,121 drives from 465 pseudonymous drivers across 115 vehicle models (975 hours), and its benchmark is built to separate a driver’s own behavior from vehicle, route and condition shortcuts. See the project page.
- BATON: A Multimodal Benchmark for Bidirectional Automation Transition Observation in Naturalistic Driving (Yuhang Wang, Yiyao Xu, Chaoyun Yang, Lingyao Li, Jingran Sun, Hao Zhou). BATON records handovers to automation and takeovers from it on the same synchronized streams — 781 routes, 173 drivers and 204.9 hours — and scores takeover both as onset detection and as pre-override anticipation. See the project page.
Both datasets are on Hugging Face, each with a small sample that is approved automatically.