List your company

Multi-Hop Question Answering on HotpotQA (test val)

47.1Exact Match (EM)DeepControl
13.61222.3063139.694Feb 2, 2026Feb 27, 2026Mar 25, 2026Apr 20, 2026May 16, 2026Jun 11, 2026Jul 7, 2026

Evaluation results

MethodLinks
DeepControlBackbone Model=Qwen2.5-7B2026.0247.147.9—
(ParaSet+SetCE)+CE5Base=Contriever, Latency (sec/query)=0.1962026.0742.9—55.3
(ParaSet+SetCE)+CE5Base=Qwen3, Latency (sec/query)=0.2412026.0741—52.9
ParaSet+SetCEBase=Contriever, Latency (sec/query)=0.1162026.0740.4—52.2
DeepControlBackbone Model=Qwen2.5-3B2026.0240.241.1—
CE10Base=Contriever, Latency (sec/query)=0.0672026.0740.1—51.6
CE10Base=Qwen3, Latency (sec/query)=0.1142026.0739.7—51.5
SetCEBase=Qwen3, Latency (sec/query)=0.5852026.0739.6—50.8
SetCEBase=Contriever, Latency (sec/query)=0.5552026.0739.1—50.6
CE5Base=Contriever, Latency (sec/query)=0.0672026.0739—50.6
ListCEBase=Contriever, Latency (sec/query)=0.5552026.0738.7—49.7
CE5Base=Qwen3, Latency (sec/query)=0.1142026.0738.4—49.4
ListCEBase=Qwen3, Latency (sec/query)=0.5852026.0738—48.6
BiBase=Qwen3, Latency (sec/query)=0.0732026.0737—47.3
ParaSet+SetCEBase=Qwen3, Latency (sec/query)=0.1642026.0736.9—48
BiBase=Contriever, Latency (sec/query)=0.0322026.0735—45.9
Direct InferenceBackbone Model=Qwen2.5-7B2026.0218.318.1—
Direct InferenceBackbone Model=Qwen2.5-3B2026.0214.913.4—