Lovish Madaan

The Art of Scaling Reinforcement Learning Compute for LLMs

Large Language Models as Improvement Operators: Better Reasoning by Iteration

Let's (not) just put things in Context: Test-time Training for Long-context LLMs

Beyond Verifiable Rewards: Scaling Reinforcement Learning in Language Models to Unverifiable Data

MLGym: A New Framework and Benchmark for Advancing AI Research Agents

Correlating and Predicting Human Evaluations of Language Models from Natural Language Processing Benchmarks

Quantifying Variance in Evaluation Benchmarks