Stanford-Led Team Wins Databricks Grounded Reasoning Cup at 63.3%

At the Data + AI Summit, Databricks hosted a live agent competition on unseen corpora and the new OfficeQA Pro V2 benchmark; a Stanford, UMass Amherst and Yale team won with 63.3% accuracy and shared their approach.

2026-08-19 ~ 2026-08-19 · 3 related posts

• First live AI agent competition held at Data + AI Summit, winners announced — matei_zaharia · 2026-08-19
• Stanford, Yale Win First Live AI Agent Competition at Data + AI Summit — CShorten30 · 2026-08-19
• Stanford Team Wins Databricks Grounded Reasoning Cup With 63.3% Accuracy Via End-to-End Agent Optimization — jefrankle · 2026-08-19