← Search

Yicheng Lin

2 accepted papers

2025

OmniDraft: A cross-vocabulary, online adaptive drafter for on-device speculative decoding

NeurIPS 2025poster

Speculative decoding generally dictates having a small, efficient draft model that is either pretrained or distilled offline to a particular target model series, for instance, Llama or Qwen models. However, within online deployment settings, there are two major challenges: 1) usage of a target model…

Cited by 0SourceScholar
2024

Breaking of Brightness Consistency in Optical Flow With a Lightweight CNN Network

RA-L 2024

The sparse optical flow method is a fundamental task in computer vision. However, its reliance on the assumption of constant environmental brightness constrains its applicability in high dynamic range (HDR) scenes. In this study, we propose a novel approach aimed at transcending image color informat

Cited by 13SourcecodeScholar