News

Aug 05, 2026 Excited to release SeedRealtime, a native audio-visual full-duplex LLM. As a core contributor, I worked on visual-based proactive interaction and duplex RL training, enabling the model to continuously watch, listen, and respond at the right moment.
Jul 01, 2026 We released the Seed2.0 Model Card, presenting a frontier model for real-world complexity. I contributed to its streaming-video and omni-modal understanding capabilities.
Apr 15, 2026 We released the Seedance 2.0 paper, presenting a unified multimodal audio-video generation model. I contributed to its omni-modal data pipeline.
Mar 21, 2026 We released the Seed1.8 Model Card, introducing a generalized agentic model for real-world scenarios. I contributed to its streaming-video understanding and proactive visual interaction capabilities.
May 23, 2025 Excited to share that I’ve joined the Multimodal Interaction & World Model team at ByteDance! Looking forward to tackling new challenges ahead. I’m open to collaborations and mentoring interns — feel free to reach out!
Jun 17, 2024 Agent-Eval-Refine won the best paper award at CVPR’24 MAR Workshop, and is accepted to COLM 2024!
Feb 26, 2024 GROUNDHOG is accepted to CVPR 2024! see you in Seattle!
Feb 15, 2024 I will join Meta Reality Labs to work with Dr. Shane Moon in Summer 2024 as a Research Scientist Intern. Hope to meet new friends in Seattle!
Oct 07, 2023 Two papers accepted to EMNLP 2023!
Jun 07, 2023 We won the First Place ($500,000) in the 1st Amazon Alexa Prize SimBot Challenge! It was an absolute honor to co-lead the amazing Team SEAGULL with Jed! Big congrats to all of our team members! 🎉 Read our technical report here.
Mar 17, 2023 I will join Amazon Alexa AI to work with Dr. Qiaozi Gao in Summer 2023 as a Research Scientist Intern. Hope to meet new friends in Sunnyvale!