Neel Jain

DynaGuard: A Dynamic Guardian Model With User-Defined Policies

Scalable Pretraining of Retrieval Models

Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach

LiveBench: A Challenging, Contamination-Limited LLM Benchmark

Refusal Tokens: A Simple Way to Calibrate Refusals in Large Language Models

Running Huge Context Windows On Tiny GPUs

Refusal Tokens: A Simple Way to Calibrate Refusals in Large Language Models

GenQA: An Instruction Dataset of LLM Generated Questions and Answers