DeltaForge: A Post-hoc Mixture-of-Experts VLA for Multi-Domain Autonomous Driving
DeltaForge: A Post-hoc Mixture-of-Experts VLA for Multi-Domain Autonomous Driving
This paper introduces DeltaForge, the first post-hoc mixture-of-experts vision-language-action (VLA) model designed for multi-domain autonomous driving.
The authors propose DeltaForge, a post-hoc mixture-of-experts framework that consolidates independently finetuned per-domain vision-language-action models into a single multi-domain model using singular value decomposition to separate shared and domain-specific weight updates.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
Source：arXiv Robotics — research abstracts · arxiv.org