List your company

Question Answering on MuSiQue (F1)

52.27F1 ScorePANINI
7.581219.183130.78542.3869Jun 4, 2024Sep 29, 2024Jan 24, 2025May 21, 2025Sep 15, 2025Jan 10, 2026May 7, 2026

Evaluation results

MethodLinks
PANINIRetrieval Category=Proposed Method2026.0252.27
OracleBackbone=Llama 3.3 70B, Context Length=1500 words2026.0151.9
OracleBackbone=Llama 3.3 70B, Context Length=1000 words2026.0149.7
HippoRAG 2Retrieval Category=Structure-Augmented RAG2026.0249.3
OracleBackbone=Qwen 2.5 72B, Context Length=1000 words2026.0148.7
OracleBackbone=Qwen 2.5 72B, Context Length=1500 words2026.0148.6
OracleBackbone=Llama 3.3 70B, Context Length=500 words2026.0146.2
NV-Embed-v2 (7B)Retrieval Category=Dense Retrieval2026.0246
IRCoTRetrieval Category=Agentic Systems2026.0244.2
OracleBackbone=Qwen 2.5 72B, Context Length=500 words2026.0143.8
Qwen3-Embedding (8B)Retrieval Category=Dense Retrieval, Reranker=Qwen3-Embedding (8B) + reranker2026.0243.7
CoABackbone=text-unicorn, Context Window=8k2024.0642.49
GraphRAGRetrieval Category=Structure-Augmented RAG2026.0242
Search-R1Retrieval Category=Agentic Systems2026.0241.1
RAGBackbone=text-unicorn, Context Window=8k2024.0640.38
Qwen3-Embedding (8B)Retrieval Category=Dense Retrieval, Reranker=None2026.0239.4
RAPTORRetrieval Category=Structure-Augmented RAG2026.0239.2
DF-RAG + ICBackbone=Llama 3.3 70B, Context Length=1000 words2026.0138.6
LongRAGBackbone=Llama 3.3 70B, Context Length=independent2026.0138.1
DF-RAG + ICBackbone=Llama 3.3 70B, Context Length=1500 words2026.0138
DF-RAGBackbone=Llama 3.3 70B, Context Length=1500 words2026.0137.9
RAG with gMMRBackbone=Llama 3.3 70B, Context Length=1000 words2026.0137.5
CoABackbone=text-bison, Context Window=8k2024.0637.09
LongRAGBackbone=Qwen 2.5 72B, Context Length=independent2026.0136.8
DF-RAGBackbone=Llama 3.3 70B, Context Length=1000 words2026.0136.3
Vanilla RAGBackbone=Llama 3.3 70B, Context Length=1000 words2026.0136
RAG with gMMRBackbone=Llama 3.3 70B, Context Length=1500 words2026.0135.9
HippoRAGRetrieval Category=Structure-Augmented RAG2026.0235.9
Vanilla RAGBackbone=Llama 3.3 70B, Context Length=1500 words2026.0135.4
CoABackbone=gemini-ultra, Context Window=8k2024.0635.09
RAG with gMMRBackbone=Qwen 2.5 72B, Context Length=1500 words2026.0134.6
RAPTORBackbone=Llama 3.3 70B, Context Length=independent2026.0134.3
DF-RAG + ICBackbone=Llama 3.3 70B, Context Length=500 words2026.0134
RAGBackbone=text-bison, Context Window=8k2024.0633.83
Vanilla RAGBackbone=Qwen 2.5 72B, Context Length=1500 words2026.0133.6
HippoRAGBackbone=Llama 3.3 70B, Context Length=independent2026.0133.5
DF-RAGBackbone=Llama 3.3 70B, Context Length=500 words2026.0133.3
RAG with gMMRBackbone=Qwen 2.5 72B, Context Length=1000 words2026.0133.3
DF-RAG + ICBackbone=Qwen 2.5 72B, Context Length=1500 words2026.0133.3
HippoRAGBackbone=Qwen 2.5 72B, Context Length=independent2026.0132.8
DF-RAG + ICBackbone=Qwen 2.5 72B, Context Length=500 words2026.0132.7
DF-RAGBackbone=Qwen 2.5 72B, Context Length=1500 words2026.0132.2
Vanilla RAGBackbone=Qwen 2.5 72B, Context Length=1000 words2026.0131.9
DF-RAG + ICBackbone=Qwen 2.5 72B, Context Length=1000 words2026.0131.7
RAGBackbone=gemini-ultra, Context Window=8k2024.0631.56
DF-RAGBackbone=Qwen 2.5 72B, Context Length=500 words2026.0131.3
RAG with gMMRBackbone=Llama 3.3 70B, Context Length=500 words2026.0130.9
Vanilla RAGBackbone=Llama 3.3 70B, Context Length=500 words2026.0129.9
BM25Retrieval Category=Sparse Retrieval, Reranker=BM25 + reranker2026.0229.9
VanillaBackbone=text-unicorn, Context Window=8k2024.0629.67
RAPTORBackbone=Qwen 2.5 72B, Context Length=independent2026.0129.6
DF-RAGBackbone=Qwen 2.5 72B, Context Length=1000 words2026.0128.8
VanillaBackbone=gemini-ultra, Context Window=32k2024.0627.93
RAG with gMMRBackbone=Qwen 2.5 72B, Context Length=500 words2026.0127.6
VanillaBackbone=text-bison, Context Window=8k2024.0626.87
TIPSBackbone=Qwen-2.5-7B Instruct2026.0326.58
BM25Retrieval Category=Sparse Retrieval, Reranker=None2026.0225.3
Vanilla RAGBackbone=Qwen 2.5 72B, Context Length=500 words2026.0124.8
VanillaBackbone=gemini-ultra, Context Window=8k2024.0623.61
INTRAGenerator=T5Gemma2026.0523
Qwen3-Emb-4B + RerankerGenerator=T5Gemma2026.0522.5
NoneRetrieval Category=No Retrieval2026.0222
Qwen3-Emb-4BGenerator=T5Gemma2026.0521.8
Hybrid RAGGenerator=T5Gemma2026.0520.1
Qwen3-Emb-0.6BGenerator=T5Gemma2026.0520
MaxSimGenerator=T5Gemma2026.0519.8
BGEGenerator=T5Gemma2026.0519.7
PPOBackbone=Qwen-2.5-7B Instruct2026.0317.79
BM25Generator=T5Gemma2026.0516.8
TIPSBackbone=Qwen-2.5-3B Instruct2026.0315.85
GRPOBackbone=Qwen-2.5-3B Instruct2026.0314.81
TF-IDFGenerator=T5Gemma2026.0514.6
MT-GRPO*Backbone=Qwen-2.5-7B Instruct2026.0314.29
PPOBackbone=Qwen-2.5-3B Instruct2026.0312.65
MT-PPOBackbone=Qwen-2.5-7B Instruct2026.0312.13
GRPOBackbone=Qwen-2.5-7B Instruct2026.0312.12
MT-PPOBackbone=Qwen-2.5-3B Instruct2026.0311.05
MT-GRPO*Backbone=Qwen-2.5-3B Instruct2026.0310.68
LightRAGRetrieval Category=Structure-Augmented RAG2026.029.3