Authors

Yuyang He

Publications 2

Preprint Open access

FastBench: Can Streaming VLMs Perceive High-Dynamic Real-World Streams?

Yuxuan Hu, Weikang Shi, Yang Bo et al. · 2026

Streaming Video Large Language Models (VLMs) enable continuous video understanding, yet existing benchmarks focus on low-dynamic scenarios. Under bounded context budgets, models must balance temporal history, spatial resolution, and temporal granularity; sparse sampling at 1--2 FPS misses fast events. We introduce Fast …

Preprint Open access

D-Loop: Looped Diffusion Drafting for Speculative Decoding

Kecheng Chen, Yuyang He, Cheng Gong et al. · 2026

Block diffusion accelerates speculative decoding by drafting multiple tokens in one forward pass. However, each position predicts a marginal distribution without observing earlier proposed tokens, limiting draft quality and acceptance length. We identify a concrete failure, the \emph{repetition trap}, in which neighbor …

Co-authors

Wenbo Li Rui Liu Yuxuan Hu Weikang Shi Yang Bo Xudong Lu Xintong Guo Shuhan Li Huankang Guan Peiwen Sun Yunqiao Yang Hongsheng Li