List your company

Multi-document Retrieval on HotpotQA, 2WikiMultihopQA, and MuSiQue (val)

64.61Average F1GritLM
28.230837.675447.1256.5646Feb 12, 2026

Evaluation results

MethodLinks
GritLMModel Category=Dense Models, top_k=32026.0264.61
GTE-Qwen2Model Category=Dense Models, top_k=32026.0262.25
AttentionRetrieverModel Category=AttentionRetriever (Ours), Backbone=LLaMA-3.2, Model size=3B, top_k=32026.0262.23
Qwen3Model Category=Dense Models, top_k=32026.0261.58
AttentionRetrieverModel Category=AttentionRetriever (Ours), Backbone=Qwen-2.5, Model size=3B, top_k=32026.0261.51
GTRModel Category=Dense Models, top_k=32026.0256.98
ANCEModel Category=Dense Models, top_k=32026.0254.78
SPScannerModel Category=Autoregressive Models, top_k=32026.0254.34
BM25Model Category=Sparse Models, top_k=32026.0248.54
DPRModel Category=Dense Models, top_k=32026.0246.96
CDEModel Category=Dense Models, top_k=32026.0229.63