ByteDance's Seed Multimodal Interaction and World Model team is seeking PhD students for a 2026 internship focused on building rl training systems for large-scale multimodal models and developing unified video-audio-language frameworks with visual latent reasoning.
Interns contribute to research, collaborate with researchers, and influence product directions. Applications are reviewed on a rolling basis; please state your start and end dates in your resume.
#J-18808-Ljbffr
PhD Intern: Multimodal RL & Video-Language Foundations
ByteDance · San Jose, CA, USA ·
- Pay:
- 68.191 - 97.120
- Job type:
- Full Time