List your company

LLM-as-a-judge on HH-RLHF

81.3CoverageHeuristic Selection
28.98842.56956.1569.731May 14, 2026

Evaluation results

MethodLinks
Heuristic SelectionCascade=M → L → O, Target agreement level=0.852026.0581.30
Heuristic SelectionCascade=L → Q → O, Target agreement level=0.852026.0572.40
CSE + PredictiveCascade=M → L → O, Target agreement level=0.852026.0572.30
CSE + PredictiveCascade=L → Q → O, Target agreement level=0.852026.0558.60
CSE + OursCascade=M → L → O, Target agreement level=0.852026.0547.242.8
CSE + RandomCascade=M → L → O, Target agreement level=0.852026.0545.429.8
CSE + SimulatedCascade=M → L → O, Target agreement level=0.852026.0542.537
CSE + VanillaCascade=M → L → O, Target agreement level=0.852026.0541.333.4
CSE + OursCascade=L → Q → O, Target agreement level=0.852026.0536.951.6
CSE + RandomCascade=L → Q → O, Target agreement level=0.852026.0534.235.7
CSE + VanillaCascade=L → Q → O, Target agreement level=0.852026.0532.737.9
CSE + SimulatedCascade=L → Q → O, Target agreement level=0.852026.053146.9