👋 Open to Work

Zhenghao Xing

harryhsing

·

https://harryhsing.github.io/

AI & ML interests

Multimodal Agents · Video Reasoning

Recent Activity

upvoted a paper 27 days ago

ReflectWorld-MM: An Entity-Oriented Multimodal Memory System for Open-Ended Video Streams

updated a Space about 1 month ago

harryhsing/omniagent-rl

updated a Space about 1 month ago

harryhsing/OmniAgent-Slides

View all activity

Organizations

Papers 4

arxiv:2606.19341

arxiv:2510.12720

arxiv:2505.17017

arxiv:2505.04623

spaces 2

OmniAgent-RL

Agentic video understanding with active perception loop

OminAgent Slides

Answer video questions by actively searching frames and audio

models 4

harryhsing/OmniAgent-SFT-7B

Video-Text-to-Text • 11B • Updated Jun 18 • 16 • 1

harryhsing/OmniAgent-RL-7B

Video-Text-to-Text • 9B • Updated Jun 18 • 41 • 1

harryhsing/EchoTraffic-7B

harryhsing/EchoInk-R1-7B

9B • Updated May 11, 2025 • 25 • 3

datasets 2

harryhsing/AV-TAU

Viewer • Updated Aug 20, 2025 • 148k • 40 • 4

harryhsing/AVQA-R1-6K

Viewer • Updated May 8, 2025 • 8.98k • 262 • 3