Harold Haodong Chen

Biography

I am a PhD student at HKUST and currently interning at Tongyi Lab@Alibaba Group, working on generation.

My current research interests focus primarily on Visual Generation/Reasoning and Code-for-Vision. Feel free to email me if you are interested in working with me :-)

Preprints/Reports

• The Past Frames the Future: Memory for Autoregressive Video Generation [arXiv]
• GenRouter: Unified Workflow Routing for Agentic Image Generation [arXiv]
• RoboEvolve: Co-Evolving Planner-Simulator for Robotic Manipulation [arXiv]
• DVD: Deterministic Video Depth Estimation with Generative Priors [arXiv]

Publications

• Show, Don't Tell: Morphing Latent Reasoning into Image Generation ICML 2026 [Paper]
• ScalingAR: Scaling Confidence for Autoregressive Image Generation ICML 2026 [Paper]
• TiViBench: Benchmarking Think-in-Video Reasoning for Video Generation CVPR 2026 [Paper]
• Hierarchical Fine-grained Preference Optimization for Physically Plausible Video Generation NeurIPS 2025 [Paper]
• FineQuest: Adaptive Knowledge-Assisted Sports Video Understanding via Agent-of-Thoughts Reasoning MM 2025 (Oral) [Paper]
• VistaDPO: Video Hierarchical Spatial-Temporal Direct Preference Optimization for Large Video Models ICML 2025 [Paper]
• SeFAR: Semi-supervised Fine-grained Action Recognition with Temporal Perturbation and Learning Stabilization AAAI 2025 [Paper]
• FineCLIPER: Multi-modal Fine-grained CLIP for Dynamic Facial Expression Recognition with AdaptERs MM 2024 [Paper]

Talks

• Jul. 2026: Deterministic Video Depth Estimation with Generative Priors @ TechBeat
• Jan. 2026: Think-in-Video Reasoning for Video Generation @ TwelveLabs

Services

• Reviewer: CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR, ACL, AAAI, MM, TMLR, Pattern Recognition.