List your company

Question Answering on Natural QA Hard

61.47F1 ScoreEND
45.027649.296353.56557.8337Feb 26, 2025

Evaluation results

MethodLinks
ENDModel=Llama3-8b-instruct2025.0261.4797
ENDModel=Qwen2.5-7b-instruct2025.0259.0795
LLM-DiscrimModel=Llama3-8b-instruct2025.0258.72—
RAGModel=Llama3-8b-instruct2025.0258.18—
RAGModel=Qwen2.5-7b-instruct2025.0255.37—
LLM-DiscrimModel=Qwen2.5-7b-instruct2025.0254.7—
RAGModel=Mistral-v0.3-instruct2025.0248.58—
ENDModel=Mistral-v0.3-instruct2025.0247.8894
LLM-DiscrimModel=Mistral-v0.3-instruct2025.0245.66—