2 papers
cs.CL2026
SG-OPD: Sign-Gated On-Policy Distillation via Sign-Consistency Gating and Phased Teacher Sampling
Haoran Xu, Hongyu Wang, Yifei Gao +3
On-policy distillation (OPD) trains a student on its own trajectories with dense per-token supervision from a stronger teacher, and often outperforms off-policy distillation and st…
cs.CV2026
Visual Para-Thinker++: A Single-Policy Multi-Agent Framework for Visual Reasoning
Haoran Xu, Hongyu Wang, Yifei Gao +4
Visual reasoning requires integrating evidence distributed across regions, attributes, and relations, making single-chain reasoning prone to early perceptual commitment and halluci…