Fetching the paper…

MLR-Bench: Evaluating AI Agents on Open-Ended Machine Learning Research · Around