List your company

Pairwise evaluator agreement with human judgment on Procthor-10K Complex 1.0 (test)

79.5Method A SceneCritic ScoreSceneCritic
67.85270.87673.976.924Apr 14, 2026

Evaluation results

MethodLinks
SceneCriticMethod A=holodeck, Method B=layoutgpt, Difficulty=Complex2026.0479.59568.3100100087.512.593.86.210087.593.810.530.77——————
SceneCriticMethod A=holodeck, Method B=layoutvlm, Difficulty=Complex2026.0479.59594956.293.86.293.86.293.893.893.893.80.7870.7870.77——————
SceneCriticMethod A=heuristic, Method B=holodeck, Difficulty=Complex2026.0479.397579.59581.218.881.218.887.512.581.281.287.50.4260.3830.563——————
SceneCriticMethod A=heuristic, Method B=layoutgpt, Difficulty=Complex2026.0479.397568.310087.512.587.512.5100087.587.51000.5620.531——————
SceneCriticMethod A=heuristic, Method B=layoutvlm, Difficulty=Complex2026.0479.3975949518.881.26.293.812.587.581.293.887.50.4260.7870.563——————
SceneCriticMethod A=layoutgpt, Method B=layoutvlm, Difficulty=Complex2026.0468.3100949501006.293.8010010093.810010.7871——————