List your company

Question Answering on HotpotQA (Cover Exact Match)

52.4Cover EMPAGER
27.64834.07440.546.926Jan 14, 2026

Evaluation results

MethodLinks
PAGERBackbone=Llama3.1-70B-Instruct2026.0152.4
DeepNoteBackbone=Llama3.1-70B-Instruct2026.0151.7
PAGERBackbone=Qwen3-32B2026.0150.6
Search-o1Backbone=Llama3.1-70B-Instruct2026.0150.2
StructRAGBackbone=Llama3.1-70B-Instruct2026.0149.6
RATBackbone=Llama3.1-70B-Instruct2026.0148.9
DeepNoteBackbone=Qwen3-32B2026.0148.4
Vanilla RAGBackbone=Llama3.1-70B-Instruct2026.0148.2
Search-o1Backbone=Qwen3-32B2026.0147.6
IRCoTBackbone=Qwen3-32B2026.0145.1
RATBackbone=Qwen3-32B2026.0143.7
Iter-RetGenBackbone=Qwen3-32B2026.0143.2
Vanilla RAGBackbone=Qwen3-32B2026.0142.9
StructRAGBackbone=Qwen3-32B2026.0142.5
Iter-RetGenBackbone=Llama3.1-70B-Instruct2026.0141.1
Vanilla LLMBackbone=Llama3.1-70B-Instruct2026.0137.7
IRCoTBackbone=Llama3.1-70B-Instruct2026.0135.2
Vanilla LLMBackbone=Qwen3-32B2026.0128.6