Back to researchers

Sandhini Agarwal

Instruction tuning and RLHF

Worked on instruction-following and RLHF practices that became the standard post-training recipe.

Highlights

OpenAIRLHFPost-training
Focus: Instruction tuning and RLHF
Why it matters: Worked on instruction-following and RLHF practices that became the standard post-training recipe.

Research Areas

OpenAIRLHFPost-training
Sandhini Agarwal - AI Researcher Profile | 500AI