BOLT: Online Lightweight Adaptation for Preparation-Free Heterogeneous Cooperative Perception
Kang Yang, Tianci Bu, Peng Wang, Deying Li, Yongcai Wang · 2026
Read BOLT: Online Lightweight Adaptation for Preparation-Free Heterogeneous Cooperative Perception by Kang Yang, Tianci Bu, Peng Wang, Deying Li, Yongcai Wang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-00405.txt.
Description
Kang Yang and 4 co-authors tackle cooperative perception for vehicles and robots whose detectors were trained separately, by different developers, with no joint preparation. In the wild, such agents meet only occasionally and online. Simply fusing their features, the authors find, performs worse than one agent perceiving alone, because the feature languages do not match.
Their module, BOLT, adapts neighbors' features on the fly using the host agent as teacher, needing no ground-truth labels. Confident host features guide the alignment, while neighbors are invited to fill the regions where the host is unsure. The module is tiny by modern standards and plugs into existing systems. On two driving benchmarks it lifted detection scores far above unadapted fusion and consistently beat working alone, across pairing and fusion choices.
Catalog record
| CREATOR | Kang Yang, Tianci Bu, Peng Wang, Deying Li, Yongcai Wang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.00405