About

Enxin Song is a research intern at the University of California, San Diego (UCSD) under Prof. Zhuowen Tu. She will receive her M.S. in March 2026 from Zhejiang University, advised by Prof. Gaoang Wang (CVNext Lab), and holds a B.S. in Software Engineering from Dalian University of Technology. Previously, she was a research intern at Microsoft Research Asia. Enxin stays curious about the wider landscape of computer vision and deep learning, and actively seeks new collaboration opportunities. Her work centers on video understanding, highlighted by MovieChat, the first Large Mutli-Modal Model for hour-long video understanding. She has co-organized workshops and challenges on video understanding at CVPR 2024 and 2025. She is a highly self-motivated, and curious student applying to Ph.D. programs for 2026Fall.

Experiences

Feb. 2025 -- Present, University of California, San Diego (UCSD), USA Visiting Intern, Advised by Prof. Zhuowen Tu.
Nov. 2023 -- May. 2024, Media Computing Group, Microsoft Research Lab - Asia, Beijing, China Research Intern, Work on text-to-image generation

Education

M.S. Sep. 2023 - Mar. 2026 (expected) Zhejiang University, Hangzhou, China. 1/82, M.S. in Artificial Intelligence
B.S. Sep. 2019 - Jun. 2023 Dalian University of Technology (DLUT) , Dalian, China. 21/385, B.S. in Software Engineering

News

Our paper MovieChat+: Question-aware Sparse Memory for Long Video Question Answering is accepted by IEEE TPAMI.
Our paper Video-MMLU: A Massive Multi-Discipline Lecture Understanding Benchmark is accepted by ICCV 2025 Findings.
Our paper Bringing RNNs Back to Efficient Open-Ended Video Understanding is accepted by ICCV 2025.
We are hosting two CVPR 2025 Video Understanding Challenge @ LOVE Track 1A and LOVE Track 1B.
We release Video-MMLU, a Massive Multi-Discipline Lecture Understanding Benchmark.
One paper accepted to CVPR 2025 workshop@Efficient Large Vision Models.
Two papers are accepted by ICLR 2025.

Selected Publications and Manuscripts

* Equal contribution.

Also see Google Scholar.

Video-MMLU: A Massive Multi-Discipline Lecture Understanding Benchmark

Enxin Song, Wenhao Chai, Weili Xu, Jianwen Xie, Yuxuan Liu, Gaoang Wang,

ICCVW, 2025

[ Paper] [Website] [Benchmark] [Code]

Video-MMLU is a massive benchmark designed to evaluate the capabilities of LMMs in understanding Multi-Discipline Lectures.

AuroraLong: Bringing RNNs Back to Efficient Open-Ended Video Understanding

Weili Xu*, Enxin Song*, Wenhao Chai, Xuexiang Wen, Tian Ye, Gaoang Wang,

ICCV, 2025

[ Paper]

Video-MMLU uses a linear RNN language model that handles input sequence of arbitrary length with constant-size hidden states to solve long video understanding tasks.

AuroraCap: Efficient, Performant Video Detailed Captioning and a New Benchmark

Wenhao Chai*, Enxin Song*, Yilun Du, Chenlin Meng, Vashisht Madhavan, Omer Bar-Tal, Jeng-Neng Hwang, Saining Xie, Christopher D. Manning

ICLR, 2025

[ Paper] [Website] [Model] [Benchmark] [Code]

AuroraCap is a multimodal LLM designed for image and video detailed captioning. We also release VDC, the first benchmark for detailed video captioning.

MovieChat: From Dense Token to Sparse Memory for Long Video Understanding

Enxin Song*, Wenhao Chai*, Guanhong Wang*, Yucheng Zhang, Haoyang Zhou, Feiyang Wu, Xun Guo, Tian Ye, Yan Lu, Jenq-Neng Hwang, Gaoang Wang

CVPR, 2024

[ Paper] [Website] [Benchmark] [Code]

MovieChat achieves state-of-the-art performace in extra long video (more than 10K frames) understanding by introducing memory mechanism.

Professional Service

Conference and Journal Refereeing:
NeurIPS 2025, PRCV 2025, CVPR 2025, ICLR 2025, TMM 2024, PRCV 2023

Workshop Organization:
Workshop on Long-form Video Understanding at CVPR 2025
Workshop on Long-form Video Understanding at CVPR 2024

Teaching Assistant

ECE 445 Senior Design (Undergraduate) - Sping 2024
Teaching Assistant (TA), with Prof. Gaoang Wang

Selected Honors & Awards

National Scholarship, 2024 (Zhejiang University)
National Scholarship, 2021 (Dalian University of Technology)

Top