HKUST-GZ UniWAM: Fusing VLA and WAM into a Single Transformer, Dual SOTA in Simulation and on Real Robots

A team at HKUST-GZ proposes UniWAM, which uses a Mixture-of-Transformers (MoT) architecture to integrate physical reasoning, future visual generation, and action prediction into one model. Built on three expert modules—Qwen3-VL-2B-Instruct, the Wan2.2-TI2V-5B backbone, and an action predictor—and paired with the EgoANT pipeline for automatic annotation of egocentric human video, it is jointly pretrained on human data, robot demonstrations, and VQA data.