MediationBench compares AI mediators on 62 simulated conflicts and labels early results as provisional.
MediationBench — testing AI mediation in simulated conflicts Skip to main content Mediation Bench Leaderboard Research Transcript Contribute About hai.ai Published by Human Assisted Intelligence, PBC MediationBench MediationBench compares AI mediators on the same 62 simulated two-party conflicts. Within each test, every system works with the same model-played parties under the same rules. Early results are labeled provisional: one complete run, scored by one or two judge models. View the leaderboard How scoring works How it works Each run covers 62 authored disputes. Language models play both
MediationBench compares AI mediators on 62 simulated conflicts and labels early results as provisional.
MediationBench — testing AI mediation in simulated conflicts Skip to main content Mediation Bench Leaderboard Research Transcript Contribute About hai.ai Published by Human Assisted Intelligence, PBC MediationBench MediationBench compares AI mediators on the same 62 simulated two-party conflicts. Within each test, every system works with the same model-played parties under the same rules. Early results are labeled provisional: one complete run, scored by one or two judge models. View the leaderboard How scoring works How it works Each run covers 62 authored disputes. Language models play both
Related listings
Just spotted: Kendall Jenner in Los Angeles and Kanye West in Los Angeles. The restaurants, bars and hotels celebrities ...