List your company

Output Diversity on NOVELTYBENCH

52.42Distinct ScoreDeepSeek-14B
0.565614.027827.4940.9522Jan 16, 2026Feb 7, 2026Mar 1, 2026Mar 23, 2026Apr 14, 2026May 6, 2026May 28, 2026

Evaluation results

MethodLinks
DeepSeek-14Bsampling_setting=S-best2026.0152.42
DeepSeek-14Bsampling_setting=Mixed2026.0152.07
Qwen3-32Bsampling_setting=Mixed2026.0146.53
DeepSeek-14Bsampling_setting=S-non-en avg2026.0146.03
Qwen3-8Bsampling_setting=Mixed2026.0143.73
Qwen3-32Bsampling_setting=S-best2026.0143.38
Qwen3-8Bsampling_setting=S-best2026.0141.33
Qwen3-32Bsampling_setting=S-non-en avg2026.0140.3
DeepSeek-14Bsampling_setting=S-en2026.0138.33
Qwen3-14Bsampling_setting=Mixed2026.0138
Qwen3-14Bsampling_setting=S-best2026.0136.87
Qwen3-8Bsampling_setting=S-non-en avg2026.0136
Qwen3-32Bsampling_setting=S-en2026.0135
Qwen3-14Bsampling_setting=S-non-en avg2026.0131.45
Qwen3-8Bsampling_setting=S-en2026.0128.55
Qwen3-14Bsampling_setting=S-en2026.0126.2
BaseModel=LLaMA-3.1-8B2026.058.79
BaseModel=Qwen3-4B2026.058.78
BaseModel=OLMo-3-7B2026.058.17
REDIPOModel=OLMo-3-7B2026.057.43
REDIPOModel=LLaMA-3.1-8B2026.057.42
REDIPOModel=Qwen3-4B2026.056
InstructModel=OLMo-3-7B2026.055.57
DivPOModel=OLMo-3-7B2026.055.23
DPOModel=OLMo-3-7B2026.055.17
InstructModel=LLaMA-3.1-8B2026.055.16
DivPOModel=LLaMA-3.1-8B2026.054.97
DPOModel=LLaMA-3.1-8B2026.053.67
DivPOModel=Qwen3-4B2026.052.57
InstructModel=Qwen3-4B2026.052.56
DPOModel=Qwen3-4B2026.052.56