List your company

Question Answering on En.QA

47.8F1Oracle
11.441620.880830.3239.7592Jan 23, 2026Feb 14, 2026Mar 9, 2026Apr 1, 2026Apr 24, 2026May 17, 2026Jun 9, 2026

Evaluation results

MethodLinks
OracleBackbone=Llama 3.3 70B, Context Length=1500 words2026.0147.8—
OracleBackbone=Qwen 2.5 72B, Context Length=1500 words2026.0147.4—
OracleBackbone=Llama 3.3 70B, Context Length=1000 words2026.0145.2—
OracleBackbone=Qwen 2.5 72B, Context Length=1000 words2026.0144.1—
OracleBackbone=Llama 3.3 70B, Context Length=500 words2026.0141.3—
OracleBackbone=Qwen 2.5 72B, Context Length=500 words2026.0140.1—
DocTrace (w/o DM)Category=Ours2026.0637.1922.79
DocTraceCategory=Ours2026.0637.1824.22
ComoRAGCategory=Hybrid Structured RAG2026.0634.5225.07
Naive RAGCategory=Flat RAG2026.0634.3424.57
DF-RAGBackbone=Qwen 2.5 72B, Context Length=1500 words2026.0134—
RAG with gMMRBackbone=Llama 3.3 70B, Context Length=1500 words2026.0133.2—
DF-RAG + ICBackbone=Qwen 2.5 72B, Context Length=1500 words2026.0133.2—
Vanilla RAGBackbone=Llama 3.3 70B, Context Length=1500 words2026.0133.1—
DF-RAGBackbone=Llama 3.3 70B, Context Length=1500 words2026.0133.1—
RAG with gMMRBackbone=Qwen 2.5 72B, Context Length=1500 words2026.0132.5—
RAPTOR+IRCoTCategory=Summary-based RAG2026.0632.0919.36
DF-RAG + ICBackbone=Llama 3.3 70B, Context Length=1500 words2026.0131.8—
DF-RAGBackbone=Llama 3.3 70B, Context Length=1000 words2026.0131.5—
Vanilla RAGBackbone=Qwen 2.5 72B, Context Length=1500 words2026.0131.3—
DF-RAGBackbone=Qwen 2.5 72B, Context Length=1000 words2026.0131.1—
RAG with gMMRBackbone=Qwen 2.5 72B, Context Length=1000 words2026.0130.3—
RAG with gMMRBackbone=Llama 3.3 70B, Context Length=1000 words2026.0130.1—
GPT-4o-miniCategory=LLM-only2026.0629.8312.82
DF-RAG + ICBackbone=Qwen 2.5 72B, Context Length=1000 words2026.0129.6—
DF-RAGBackbone=Llama 3.3 70B, Context Length=500 words2026.0129.3—
DF-RAG + ICBackbone=Llama 3.3 70B, Context Length=1000 words2026.0129.3—
HippoRAG2+IRCoTCategory=Graph-based RAG2026.0629.2718.24
Vanilla RAGBackbone=Llama 3.3 70B, Context Length=1000 words2026.0128.3—
DF-RAGBackbone=Qwen 2.5 72B, Context Length=500 words2026.0128.2—
RAG with gMMRBackbone=Qwen 2.5 72B, Context Length=500 words2026.0127.7—
DF-RAG + ICBackbone=Qwen 2.5 72B, Context Length=500 words2026.0127.7—
RAG with gMMRBackbone=Llama 3.3 70B, Context Length=500 words2026.0127.6—
DF-RAG + ICBackbone=Llama 3.3 70B, Context Length=500 words2026.0127.5—
Vanilla RAGBackbone=Qwen 2.5 72B, Context Length=1000 words2026.0126.9—
RAPTORCategory=Summary-based RAG2026.0626.3319.65
Vanilla RAGBackbone=Llama 3.3 70B, Context Length=500 words2026.0126.3—
Vanilla RAGBackbone=Qwen 2.5 72B, Context Length=500 words2026.0126.1—
HippoRAG2Category=Graph-based RAG2026.0624.4517.09
LongRAGBackbone=Llama 3.3 70B, Context Length=independent2026.0122.9—
LongRAGBackbone=Qwen 2.5 72B, Context Length=independent2026.0122.4—
Self-RAGCategory=Flat RAG2026.0612.844.27