List your company

Question Answering on WebQ (EM, ROUGE, F1)

32EMGRIP w/o RL
6.72813.28919.8526.411Apr 13, 2026

Evaluation results

MethodLinks
GRIP w/o RLTraining status=Training-based, Backbone=LLaMA3-8B, Retriever=BM252026.043239.935.1
GRIPTraining status=Training-based, Backbone=LLaMA3-8B, Retriever=BM252026.0431.439.334.6
DRAGINTraining status=Training-free2026.0425.231.535.7
FLARETraining status=Training-free2026.0424.230.334.7
GPT-4oTraining status=Training-free2026.0423.532.337
R1-SearcherTraining status=Training-based, Backbone=LLaMA3-8B, Retriever=BM252026.0421.826.130.6
RetRobustTraining status=Training-based, Backbone=LLaMA3-8B, Retriever=BM252026.0421.827.431.7
RobustRAGTraining status=Training-based, Backbone=LLaMA3-8B, Retriever=BM252026.0421.52529.1
Robust-RAGBackbone=Qwen2.5-7B-Instruct2026.0421.125.830
Robust-RAGBackbone=Qwen-3-4B-Instruct2026.0419.824.528.7
InstructRAGTraining status=Training-based, Backbone=LLaMA3-8B, Retriever=BM252026.0419.526.831.4
R1-SearcherBackbone=Qwen2.5-7B-Instruct2026.0419.326.730.7
GRIPBackbone=Qwen2.5-7B-Instruct2026.0419.328.333.2
ETCTraining status=Training-free2026.0418.926.630.4
SFT-RAGTraining status=Training-based, Backbone=LLaMA3-8B, Retriever=BM252026.0418.918.323.1
INFO-RAGTraining status=Training-based, Backbone=LLaMA3-8B, Retriever=BM252026.0418.127.728.9
GRIPBackbone=Qwen-3-4B-Instruct2026.0416.827.532.3
GainRAGTraining status=Training-based, Backbone=LLaMA3-8B, Retriever=BM252026.0416.524.528.9
Self-RAGTraining status=Training-based, Backbone=LLaMA3-8B, Retriever=BM252026.0416.426.527.4
GPT-3.5 TurboTraining status=Training-free2026.0415.927.431.9
Single-RAGBackbone=Qwen2.5-7B-Instruct2026.0415.322.626.3
InstructTraining status=Training-free2026.0414.825.129.4
InstructBackbone=Qwen2.5-7B-Instruct2026.0414.522.626.4
Single RAGTraining status=Training-free2026.041422.726.6
R1-SearcherBackbone=Qwen-3-4B-Instruct2026.0412.921.625.4
Single-RAGBackbone=Qwen-3-4B-Instruct2026.048.716.720.2
InstructBackbone=Qwen-3-4B-Instruct2026.047.716.320