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

Output Diversity on MT-Bench

48.36Lexical Diversity Score

QEMPO-KL

Updated 1mo ago

Evaluation Results

Method	Links
QEMPO-KL 2026.02		48.36	30.55	53.38	44.1
SPL 2026.02		47.88	30.83	52.18	43.63
QEMPO 2026.02		46.51	30.84	53.95	43.77
QEMPO 2026.02		44.23	30.18	54.29	42.9
QEMPO-KL 2026.02		44.18	30.08	53.3	42.52
RLHF 2026.02		43.85	29.08	52.43	41.79
SPL 2026.02		43.53	30.09	54.52	42.71
Qwen2.5-1.5B-Instruct 2026.02		41.27	29.2	53.39	41.29
Llama-3.2-1B-Instruct 2026.02		40.78	29.54	53.55	41.29
RLHF 2026.02		40.42	28.74	51.92	40.36
SPL 2026.02		36.18	30.31	55.77	40.75
QEMPO 2026.02		35.6	28.7	53.42	39.24
QEMPO 2026.02		35.37	28.34	52.45	38.72
Llama-3.1-8B-Instruct 2026.02		34.85	29.55	54.12	39.51
QEMPO-KL 2026.02		34.6	28.08	52.34	38.34
RLHF 2026.02		34.5	27.73	51.85	38.03
QEMPO-KL 2026.02		33.63	30.68	54.91	39.74
RLHF 2026.02		31.98	27.41	51.48	36.95
SPL 2026.02		31.81	27.41	52.34	37.19
Qwen2.5-7B-Instruct 2026.02		31.68	28.04	52.28	37.33