WAM-Diff2 Distills Autoregressive Autonomous Driving VLA into Diffusion for 15.1x Faster Decoding

机器之心 · wechat · 2026-08-19

Fudan University and Yinwang propose WAM-Diff2, a general pipeline that smoothly converts a mature multi-task autoregressive VLA into a discrete diffusion architecture, addressing two bottlenecks in autonomous driving VLAs: linearly growing serial decoding latency and exposure bias from teacher forcing. Built on a Qwen3-VL backbone:

Achieves 91.1 PDMS on NAVSIM, beating specialized planners like ReCogDrive (90.8); 49.55% closed-loop success on Bench2Drive; long-horizon waypoint L2 error reduced 5.8% versus the AR baseline, suppressing error accumulation over the prediction horizon. Limitations: token quantization error and a student capability ceiling set by the AR teacher.

Original post →

More from Research

Research channel →