Senior Research Scientist @ NVIDIA
wenliang.dai.1995@gmail.com
I am a Senior Research Scientist at NVIDIA, working on LLM post-training with a focus on agents, reasoning, and multimodality. Previously, I was a Research Intern at Salesforce Research. I received my Ph.D. from the Hong Kong University of Science and Technology (HKUST) in 2023, advised by Prof. Pascale Fung. Outside of work, I enjoy coffee, hiking, and interesting gadgets, and I have a lovely dog.
Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
Cosmos 3: Omnimodal World Models for Physical AI
Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation
Nemotron-Cascade: Scaling Cascaded Reinforcement Learning for General-Purpose Reasoning Models
InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning
NVLM: Open Frontier-Class Multimodal LLMs
A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity
Survey of Hallucination in Natural Language Generation
Core contributor to multiple post-training efforts, including NVLM, Nemotron-Cascade, Nemotron-Cascade-2, and Nemotron 3 Ultra, with hands-on experience training models ranging from 7B to 550B parameters.
Multimodal LLMs; trained one of the first well-established instruction-tuned VLMs (InstructBLIP).
WeChat DevTools and Mini-programs.