List your company

Question Answering on WikiQA (F1/Acc)

59AccuracyTS (C)
11.99224.19636.448.604Sep 30, 2025Oct 24, 2025Nov 17, 2025Dec 12, 2025Jan 5, 2026Jan 29, 2026Feb 23, 2026

Evaluation results

MethodLinks
TS (C)Algorithm Family=Contextual Linear2026.0259—
No-RewriteAlgorithm Family=Base2026.0249.8—
Oracle2025.0939.810.5
OursBackbone=Qwen2.52025.0930.29.7
Ours-cBackbone=Qwen2.52025.09298.82
OursBackbone=LLaMA-32025.0928.89.87
Ours-cBackbone=LLaMA-32025.0928.29.61
Hybrid2025.09288.99
OursBackbone=GPT-4o2025.0926.29.3
Score AgentBackbone=LLaMA-32025.09269.35
Agent-BasedBackbone=Qwen2.52025.0925.88.84
Ours-cBackbone=GPT-4o2025.09259.15
Score AgentBackbone=GPT-4o2025.0924.89.19
Ours-cBackbone=GPT-4.12025.0924.69.15
OursBackbone=GPT-4.12025.0924.29.54
Rule AgentBackbone=Qwen2.52025.09248.15
Score AgentBackbone=Qwen2.52025.09248.55
Doc2025.0923.66.79
Rule AgentBackbone=LLaMA-32025.0923.48.12
Rule AgentBackbone=GPT-4o2025.0922.88.68
Rule AgentBackbone=GPT-4.12025.0922.89.13
Agent-BasedBackbone=GPT-4.12025.0921.49.05
Score AgentBackbone=GPT-4.12025.0920.89.15
Agent-BasedBackbone=GPT-4o2025.0920.29.11
DB2025.0918.85.13
Rule-Based2025.0918.68.68
Adaptive-RAG2025.0915.89.4
Agent-BasedBackbone=LLaMA-32025.0915.29.11
Basic QA2025.0913.84.32