Share your thoughts, 1 month free Claude Pro on usSee more

Ranking Correlation Analysis on LAM benchmark

0.879Kendall Correlation

Combined Embedding

Updated 1mo ago

Evaluation Results

Method
Combined Embedding 2026.04	0.879	0.826	49	350
Anchor Points 2026.04	0.861	0.816	55	-
Random Sampling 2026.04	0.848	0.751	127	500
Combined Embedding 2026.04	0.844	0.826	49	350
Difficulty-based 2026.04	0.841	0.766	93	500
Random-Sampling-Learn 2026.04	0.829	0.72	158	900
IRT-based 2026.04	0.821	0.736	142	800
Random-Search-Learn 2026.04	0.819	0.728	141	-
Anchor Points 2026.04	0.818	0.816	55	-
Semantic Embedding 2026.04	0.802	0.723	99	-
Combined Embedding 2026.04	0.802	0.826	49	350
Semantic Embedding 2026.04	0.8	0.723	99	-
Acoustic Embedding 2026.04	0.797	0.709	166	-
Anchor Points 2026.04	0.79	0.816	55	-
Difficulty-based 2026.04	0.775	0.766	93	500
Random Sampling 2026.04	0.771	0.751	127	500
IRT-based 2026.04	0.768	0.736	142	800
Anchor Points 2026.04	0.762	0.816	55	-
Acoustic Embedding 2026.04	0.758	0.709	166	-
Random-Sampling-Learn 2026.04	0.75	0.72	158	900
Random-Search-Learn 2026.04	0.745	0.728	141	-
Anchor Points 2026.04	0.731	0.816	55	-
Combined Embedding 2026.04	0.722	0.826	49	350
Semantic Embedding 2026.04	0.719	0.723	99	-
Acoustic Embedding 2026.04	0.719	0.709	166	-
Difficulty-based 2026.04	0.704	0.766	93	500
IRT-based 2026.04	0.696	0.736	142	800
Random Sampling 2026.04	0.692	0.751	127	500
Combined Embedding 2026.04	0.68	0.826	49	350
Anchor Points 2026.04	0.659	0.816	55	-
Random-Search-Learn 2026.04	0.651	0.728	141	-
Variance-based 2026.04	0.648	0.591	-	-
Difficulty-based 2026.04	0.643	0.766	93	500
Random-Sampling-Learn 2026.04	0.637	0.72	158	900
Random Sampling 2026.04	0.623	0.751	127	500
Semantic Embedding 2026.04	0.619	0.723	99	-
IRT-based 2026.04	0.611	0.736	142	800
Variance-based 2026.04	0.601	0.591	-	-
Difficulty-based 2026.04	0.595	0.766	93	500
Random-Search-Learn 2026.04	0.582	0.728	141	-
Random-Sampling-Learn 2026.04	0.576	0.72	158	900
Acoustic Embedding 2026.04	0.557	0.709	166	-
Variance-based 2026.04	0.555	0.591	-	-
Random Sampling 2026.04	0.544	0.751	127	500
IRT-based 2026.04	0.544	0.736	142	800
Random-Search-Learn 2026.04	0.54	0.728	141	-
Random-Sampling-Learn 2026.04	0.528	0.72	158	900
Acoustic Embedding 2026.04	0.528	0.709	166	-
Combined Embedding 2026.04	0.523	0.826	49	350
Variance-based 2026.04	0.521	0.591	-	-
Variance-based 2026.04	0.487	0.591	-	-
Random-Search-Learn 2026.04	0.48	0.728	141	-
Difficulty-based 2026.04	0.465	0.766	93	500
Semantic Embedding 2026.04	0.46	0.723	99	-
Random-Sampling-Learn 2026.04	0.428	0.72	158	900
Random Sampling 2026.04	0.427	0.751	127	500
Variance-based 2026.04	0.415	0.591	-	-
Semantic Embedding 2026.04	0.371	0.723	99	-
IRT-based 2026.04	0.364	0.736	142	800
Acoustic Embedding 2026.04	0.33	0.709	166	-