Contributed to research on SAFE (Search-augmented LLM autorater) and the F1@K metric to evaluate how reliably LLMs generate supported facts in long responses.
Post-training data, RL recipe and scaling, reward models, and inference-time techniques. Key Contributions: Core developer behind Gemini 1.0, 1.5, 2.0, 2.5, and 3.0. Earlier work on Bard and LaMDA models. chenkai kuang
You might wonder, "Why does academic research matter to me?" Earlier work on Bard and LaMDA models
If you haven’t discovered Chenkai Kuang yet, take five minutes today to look closely at his work. You might just see yourself staring back. Chenkai Kuang is an AI researcher and Senior
Chenkai Kuang is an AI researcher and Senior Staff Software Engineer at Google DeepMind. He is a core contributor to the Gemini family of multimodal models. His work focuses on post-training, reinforcement learning (RL) scaling, and reasoning capabilities. Professional Profile
Chenkai Kuang is a prominent figure in the field of , recognized primarily for his contributions as a researcher and engineer at Google DeepMind . His work focuses on the development of advanced Large Language Models (LLMs) and multimodal AI systems that are shaping the future of machine learning. Contributions to the Gemini Family