CocoaBench: Evaluating Unified Digital Agents in the Wild
Paper • 2604.11201 • Published • 36
Tommaso Cerruti
Cerru02
AI & ML interests
AI safety and evaluation
Recent Activity
new activity about 16 hours ago
evaleval/EEE_datastore:Fix LLM Stats provenance relationships upvoted an article 6 days ago
Safety Evals Should Project Test-Time Compute published an article 6 days ago
Safety Evals Should Project Test-Time ComputeOrganizations
authored a
paper 17 days ago