List your company

Question Answering on NQ, TriviaQA, PopQA, HotpotQA, 2wiki, Musique, Bamboogle, FictionalHot

37.7Exact Match (EM)ReSeek
-0.1569.67219.529.328Oct 1, 2025

Evaluation results

MethodLinks
ReSeekBackbone=Qwen2.5-7b-Instruct2025.1037.7
ZeroSearchBackbone=Qwen2.5-7b-Instruct2025.1034.6
Search-R1Backbone=Qwen2.5-7b-Instruct2025.1034.2
ReSeekBackbone=Qwen2.5-3b-Instruct2025.1031.2
Search-R1Backbone=Qwen2.5-3b-Instruct2025.1028.8
ZeroSearchBackbone=Qwen2.5-3b-Instruct2025.1028.1
RAGBackbone=Qwen2.5-7b-Instruct2025.1026.7
R1Backbone=Qwen2.5-7b-Instruct2025.1023.8
RAGBackbone=Qwen2.5-3b-Instruct2025.1023.7
R1Backbone=Qwen2.5-3b-Instruct2025.1019.6
Search-o1Backbone=Qwen2.5-3b-Instruct2025.1018.7
Search-o1Backbone=Qwen2.5-7b-Instruct2025.1018.3
Direct InferenceBackbone=Qwen2.5-7b-Instruct2025.1015.8
Direct InferenceBackbone=Qwen2.5-3b-Instruct2025.1011.8
CoTBackbone=Qwen2.5-7b-Instruct2025.109.3
CoTBackbone=Qwen2.5-3b-Instruct2025.101.3