Student Researcher (Seed – Multimodal Interaction & World Model - RL Focused) – 2026 Start (PhD)

Join the Seed Multimodal Interaction and World Model team to help develop multimodal models and future multimodal assistant products. You’ll design and implement reinforcement learning (RL) training systems for large-scale foundation models, build unified video/audio/language modeling frameworks, and explore RL methods for visual reasoning. You’ll also work with researchers to evaluate world modeling and instruction-conditioned generation tasks.