List your company

Question Answering on BIGBENCH II

100True WS ScorePromptCOS
7.4431.4755.579.53Sep 3, 2025

Evaluation results

MethodLinks
PromptCOSModel=TinyLlama-chat2025.09100200.64
PromptCOSModel=Deepseek-d-qwen2025.09100240.64
PCGModel=Gemma2-it2025.09100220.71
PromptCOSModel=Gemma2-it2025.0995290.6
PRModel=TinyLlama-chat2025.098375-0.01
PRModel=Gemma2-it2025.0981710.05
PRModel=Deepseek-d-qwen2025.0973490.13
PCGModel=TinyLlama-chat2025.0967540.06
PC*Model=Gemma2-it2025.096520.61
PC*Model=TinyLlama-chat2025.096480.52
PC*Model=Deepseek-d-qwen2025.095850.51
PCGModel=Deepseek-d-qwen2025.091130.02