Question Answering On Nq Beir

nDCG@10

评测结果

各个模型在此基准测试上的表现结果

比较表格

模型名称	nDCG@10
blended-rag-improving-rag-retriever-augmented	0.67
sgpt-gpt-sentence-embeddings-for-semantic	0.524
sgpt-gpt-sentence-embeddings-for-semantic	0.401
beir-a-heterogenous-benchmark-for-zero-shot	0.524
beir-a-heterogenous-benchmark-for-zero-shot	0.533
no-parameter-left-behind-how-distillation-and	0.633