Structured Taxonomy and Evaluation for Professional Video Generation

The length of software tasks AIs can complete has doubled every 7 months since 2019

Machine learning researchers must urgently work with policymakers to address growing risks from embodied AI by plugging gaps in existing frameworks.

We introduce ALE-bench, a new benchmark for evaluating AI systems on score-based algorithmic programming contests.

We introduce FOReCAst, a benchmark for evaluating both the prediction and confidence of models across diverse real-world forecasting tasks.