Researcher — paper.dinq.me

2026 顶会论文

作者位次 #3/5CVPR 2026Poster· Computer Vision & Pattern Recognition

ParallelVLM: Lossless Video-LLM Acceleration with Visual Alignment Aware Parallel Speculative Decoding

Quan Kong · Yuhao Shen · Yicheng Ji · Huan Li · Cong Wang

多模态对齐与偏好效率与系统OpenReview 项目页

一作ACL 2026Poster· Natural Language Processing

See the Forest for the Trees: Loosely Speculative Decoding via Visual-Semantic Guidance for Efficient Inference of Video LLM s

Yicheng Ji · Jun Zhang · Jinpeng Chen · Cong Wang · Lidan Shou · Gang Chen · Huan Li

摘要

Video Large Language Models (Video-LLMs) excel in video understanding but suffer from high inference latency due to autoregressive generation. Speculative Decoding (SD) mitigates this by applying a draft-and-verify paradigm, yet existing methods are constrained by rigid exact-match rules, severel...

效率与系统多模态代码OpenReview 项目页

作者位次 #2/10ACL 2026Findings· Findings of ACL

Efficient Inference for Large Vision-Language Models: Bottlenecks, Techniques, and Prospects

Jun Zhang · Yicheng Ji · Feiyang Ren · Yihang Li · Bowen Zeng · Zonghao Chen · Ke Chen · Lidan Shou · Gang Chen · Huan Li

摘要

Large Vision-Language Models (LVLMs) enable sophisticated reasoning over images and videos, yet their inference is hindered by a systemic efficiency barrier known as visual token dominance. This overhead is driven by a multi-regime interplay between high-resolution feature extraction, quadratic a...

多模态效率与系统代码OpenReview 项目页

Yicheng Ji

获取完整研究档案

联系方式

2026 顶会论文

ParallelVLM: Lossless Video-LLM Acceleration with Visual Alignment Aware Parallel Speculative Decoding

See the Forest for the Trees: Loosely Speculative Decoding via Visual-Semantic Guidance for Efficient Inference of Video LLM s

Efficient Inference for Large Vision-Language Models: Bottlenecks, Techniques, and Prospects

常见合作者