2026
Omni-Diffusion-Distill: Few-Step Distillation of Unified Multimodal Diffusion Large Language Models
Hong Huang; Chenhongyi Yang; Junzhe Sun; Animesh Sinha; Wuyang Chen; Yifan Jiang.
ArXiv 2026
One few-step student for every task of a unified multimodal diffusion LLM, generating images in 8 steps instead of 128.
Omni-Diffusion-Distill: Few-Step Distillation of Unified Multimodal Diffusion Large Language Models
Hong Huang; Chenhongyi Yang; Junzhe Sun; Animesh Sinha; Wuyang Chen; Yifan Jiang.
One few-step student for every task of a unified multimodal diffusion LLM, generating images in 8 steps instead of 128.
Fluid-Gen-Zero: Grounding Pretrained Video Generators in Physics without Training
Hong Huang; Yuqiu Liu; Chenyu You; Daniel Martin; Chuhang Zou; Wuyang Chen.
ArXiv 2026
Training-free framework that grounds pretrained video generators in fluid physics via simulation-guided planning and latent guidance.
Fluid-Gen-Zero: Grounding Pretrained Video Generators in Physics without Training
Hong Huang; Yuqiu Liu; Chenyu You; Daniel Martin; Chuhang Zou; Wuyang Chen.
Training-free framework that grounds pretrained video generators in fluid physics via simulation-guided planning and latent guidance.
The Collaboration Tax: How Much LLM Multi-Agent Systems Pay to Coordinate
Weixiang Sun; Zehong Wang; Hong Huang; Colby Nelson; Yijun Ma; Yanfang Ye.
Conference on Empirical Methods in Natural Language Processing (EMNLP) 2026
Measures the performance LLM multi-agent systems lose when two agents must coordinate rather than act alone, and traces it to a four-stage conversational cascade.
The Collaboration Tax: How Much LLM Multi-Agent Systems Pay to Coordinate
Weixiang Sun; Zehong Wang; Hong Huang; Colby Nelson; Yijun Ma; Yanfang Ye.
Measures the performance LLM multi-agent systems lose when two agents must coordinate rather than act alone, and traces it to a four-stage conversational cascade.
2025
SafeWork-R1: Coevolving Safety and Intelligence under the AI-45° Law
Shanghai AI Lab.
Technical Report 2025
Shanghai AI Lab technical report on coevolving safety and intelligence under the AI-45° Law.
SafeWork-R1: Coevolving Safety and Intelligence under the AI-45° Law
Shanghai AI Lab.
Shanghai AI Lab technical report on coevolving safety and intelligence under the AI-45° Law.
Frontier AI Risk Management Framework in Practice: A Risk Analysis Technical Report
Shanghai AI lab.
Technical Report 2025
Shanghai AI Lab technical report on frontier AI risk management framework in practice.
Frontier AI Risk Management Framework in Practice: A Risk Analysis Technical Report
Shanghai AI lab.
Shanghai AI Lab technical report on frontier AI risk management framework in practice.
Non-Overlap-Aware Egocentric Pose Estimation for Collaborative Perception in Connected Autonomy
Hong Huang; Dongkuan Xu; Hao Zhang; Peng Gao.
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2025 Oral
Non-overlap-aware egocentric pose estimation for collaborative perception in connected autonomy.
Non-Overlap-Aware Egocentric Pose Estimation for Collaborative Perception in Connected Autonomy
Hong Huang; Dongkuan Xu; Hao Zhang; Peng Gao.
Oral
Non-overlap-aware egocentric pose estimation for collaborative perception in connected autonomy.
IQE-CLIP: Instance-aware Query Embedding for Zero-/Few-shot Anomaly Detection in Medical Domain
Hong Huang; Weixiang Sun; Zhijian Wu; Donghuan Lu; Xian Wu; Yefeng Zheng.
IEEE Transactions on Image Processing 2026
Instance-aware query embedding for zero-/few-shot anomaly detection in medical domain.
IQE-CLIP: Instance-aware Query Embedding for Zero-/Few-shot Anomaly Detection in Medical Domain
Hong Huang; Weixiang Sun; Zhijian Wu; Donghuan Lu; Xian Wu; Yefeng Zheng.
Instance-aware query embedding for zero-/few-shot anomaly detection in medical domain.
Hybrid Neural-MPM for Interactive Fluid Simulations in Real-Time
Hong Huang*; Jingxuan Xu*; Chuhang Zou; Manolis Savva; Yunchao Wei; Wuyang Chen. (* equal contribution)
ArXiv 2025
Hybrid neural-MPM approach for interactive fluid simulations in real-time.
Hybrid Neural-MPM for Interactive Fluid Simulations in Real-Time
Hong Huang*; Jingxuan Xu*; Chuhang Zou; Manolis Savva; Yunchao Wei; Wuyang Chen. (* equal contribution)
Hybrid neural-MPM approach for interactive fluid simulations in real-time.
2024
Style-Consistent 3D Indoor Scene Synthesis with Decoupled Objects
Yunfan Zhang; Hong Huang; Zhiwei Xiong; Zhiqi Shen; Guosheng Lin; Hao Wang; Chan Hua Vun.
2025 International Joint Conference on Neural Networks (IJCNN) 2025
Style-consistent 3D indoor scene synthesis with decoupled objects.
Style-Consistent 3D Indoor Scene Synthesis with Decoupled Objects
Yunfan Zhang; Hong Huang; Zhiwei Xiong; Zhiqi Shen; Guosheng Lin; Hao Wang; Chan Hua Vun.
Style-consistent 3D indoor scene synthesis with decoupled objects.