Scholar
Yuqi Liu
Google Scholar ID: X-OlO2gAAAAJ
The Chinese University of Hong Kong
multi-modal
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
286
H-index
5
i10-index
3
Publications
6
Co-authors
10
list available
Contact
Email
yqliu24@link.cuhk.edu.hk
GitHub
Open ↗
Publications
4 items
Recoverability Is a Subspace Property: A Benchmark for Certified State Estimation from Partial PDE Observations
2026
Cited
0
MemVLN: Episodic and Procedural Memory for Vision-and-Language Navigation
2026
Cited
0
OmniGameArena: A Unified UE5 Benchmark for VLM Game Agents with Improvement Dynamics
2026
Cited
0
VP-VLA: Visual Prompting as an Interface for Vision-Language-Action Models
2026
Cited
0
Resume
Academic Achievements
- Paper 'Lyra: An Efficient and Speech-Centric Framework for Omni-Cognition' accepted by ICCV 2025
- Paper 'Reversed in Time: A Novel Temporal-Emphasized Benchmark for Cross-Modal Video-Text Retrieval' accepted by ACM MM 2024
- Paper 'UI-Ins: Enhancing GUI Grounding with Multi-Perspective Instruction-as-Reasoning'
- Paper 'ViSurf: Visual Supervised-and-Reinforcement Fine-Tuning for Large Vision-and-Language Models'
- Paper 'VisionReasoner: Unified Visual Perception and Reasoning via Reinforcement Learning'
- Paper 'Seg-Zero: Reasoning-Chain Guided Segmentation via Cognitive Reinforcement'
- Ranked 1st in Trecvid 2022 VTT task
- One paper accepted by AAAI 2023
- One paper accepted by ECCV 2022
Research Experience
- Released projects such as ViSurf, UI-Ins
- Participated in multiple research projects including VisionReasoner, Seg-Zero, etc.
Education
- The Chinese University of Hong Kong, PhD student, Supervisors: Professor JIA Jiaya, Professor YU Bei
- AIM3 Lab, Renmin University of China, Master's degree, Supervisor: Professor JIN Qin
- South China University of Technology, Bachelor's degree, Graduated in 2021
Background
Research interests include Computer Vision and Multi-modal Large Language Models. Currently a PhD student at The Chinese University of Hong Kong.
Miscellany
Personal interests not mentioned
Co-authors
8 total
Jiaya Jia
Chair Professor, HKUST; Adjunct Prof., CUHK
Zhisheng Zhong
The Chinese University of Hong Kong
Qin Jin
中国人民大学信息学院
Bohao Peng
The Chinese University of Hong Kong
Bei Yu
Professor, The Chinese University of Hong Kong
Zihao Yue
Renmin University of China
Chengyao Wang
The Chinese University of Hong Kong
Shu Liu
SmartMore Ltd.