Benchmarks to faithfully assess the real-world utility of frontier AI (open-sourced mostly as Harbor RL Environments)
AI & ML interests
Untangling the science of frontier data and evaluation
Recent Activity
View all activity
datasets 7
handshake-ai-research/bankertoolbench
Viewer • Updated • 100 • 2.2k • 10
handshake-ai-research/ATLAS-Finance
Viewer • Updated • 100 • 1.01k • 6
handshake-ai-research/cue-annotations
Viewer • Updated • 533k • 1.62k
handshake-ai-research/cue-example-pool
Viewer • Updated • 161k • 108
handshake-ai-research/studentbench
Viewer • Updated • 54 • 3.93k • 4
handshake-ai-research/VIALS
Preview • Updated • 395 • 1
handshake-ai-research/CAREBench
Preview • Updated • 70 • 4