Honglu Zhou
Scholar

Honglu Zhou

Google Scholar ID: U-Uzcs8AAAAJ
Salesforce AI Research
Video UnderstandingMultimodal and Generative AIMachine Reasoning
Citations & Impact
All-time
Citations
869
 
H-index
12
 
i10-index
15
 
Publications
20
 
Co-authors
35
list available
Resume (English only)
Academic Achievements
  • Publications: ViUniT: Visual Unit Tests for More Robust Visual Programming accepted by CVPR 2025; Contra4 accepted by EMNLP 2025; MERV accepted by ICML 2025; xGen-MM-Vid (BLIP-3-Video) released; Domain-Guided Weight Modulation for Semi-Supervised Domain Generalization accepted by WACV 2025; xGen-VideoSyn-1: High-fidelity Text-to-Video Synthesis with Compressed Representations released; xGen-MM (BLIP-3): A Family of Open Large Multimodal Models released; Organized the NeurIPS 2025 Multimodal Algorithmic Reasoning Workshop.
Research Experience
  • Internships at Salesforce AI Research (Mentors: Juan Carlos Niebles and Roberto Martín-Martín), NEC Labs (Mentors: Asim Kadav and Farley Lai), and Google YouTube (Mentors: Wei-Hong Chuang and Hassan Akbari). Collaborated closely with DeepMind and Google Research.
Education
  • Ph.D. in Computer Science from Rutgers University in 2023, supervised by Professor Mubbasir Kapadia; Bachelor of Engineering in Computer Science and Technology from Communication University of China in 2017; Bachelor of Arts in TV Editing and Directing (Post-production of Television) from Communication University of China in 2016.
Background
  • Research Interests: Multimodal and Generative AI, Video Understanding, Embodiment and Robotics. Currently a Research Scientist at Salesforce AI Research, previously worked in the Machine Learning Department at NEC Laboratories America, Inc.
Miscellany
  • Personal Interests: Open to academic collaboration