List your company

Question Answering on SQuAD golden

96.6AccuracyFaithfulRAG
80.79284.8968993.104Jun 10, 2025

Evaluation results

MethodLinks
FaithfulRAGBackbone=LLaMA 3.1-8B-Instruct2025.0696.6
Opin (Instr)Backbone=LLaMA 3.1-8B-Instruct, Method Category=Context-faithful Prompting2025.0696
OriginBackbone=LLaMA 3.1-8B-Instruct2025.0695.2
COIECDBackbone=LLaMA 3.1-8B-Instruct, Method Category=Context-faithful Decoding2025.0695.1
ATTRBackbone=LLaMA 3.1-8B-Instruct, Method Category=Context-faithful Prompting2025.0694.2
CADBackbone=LLaMA 3.1-8B-Instruct, Method Category=Context-faithful Decoding2025.0691.1
KREBackbone=LLaMA 3.1-8B-Instruct, Method Category=Context-faithful Prompting2025.0681.4