Hunyuan Multimodal Reinforcement Learning (RL) Research Intern 107383
Core
Research on reinforcement learning (RL) algorithms for multimodal models, including diffusion and autoregressive models for image, video, and 3D generation.
Role type
Research intern (PhD level)
Builds
RL infrastructure and reward modeling strategies for large-scale training of multimodal models
Domain
Artificial Intelligence / Multimodal Learning / Reinforcement Learning
Deliverable
research
Required skills
Deep learning system implementation, model training and inference optimization, CPU/GPU acceleration, distributed training and inference, RL algorithm design, diffusion models, autoregressive models
Preferred skills
Text-to-image or text-to-video generation experience, ACM/NOIP participation
Technologies
Diffusion models, Autoregressive models, RL frameworks
Responsibilities
Design and develop RL infrastructure to enable efficient large-scale training, mitigate reward hacking, and explore next-generation RL paradigms
Seniority
Intern (PhD student)
