Retrieval on Musique
28.91F1 ScoreHippoRAG2
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| HippoRAG2KG Constructor=Qwen3-235B, Retrieval Paradigm=Graph-based Text2025.10 | 28.91 | — | 48.13 | — | |
| HippoRAG2KG Constructor=GPT-5-mini, Retrieval Paradigm=Graph-based Text2025.10 | 28.8 | — | 49 | — | |
| HippoRAGKG Constructor=GPT-5-mini, Retrieval Paradigm=Graph-based Text2025.10 | 28.7 | — | 47.8 | — | |
| HippoRAG2KG Constructor=gemini-2.5-flash-lite, Retrieval Paradigm=Graph-based Text2025.10 | 28.34 | — | 47.81 | — | |
| HippoRAGKG Constructor=Llama-3.3-70B, Retrieval Paradigm=Graph-based Text2025.10 | 28.32 | — | 47.41 | — | |
| HippoRAG2KG Constructor=DeepSeek-V3, Retrieval Paradigm=Graph-based Text2025.10 | 28.14 | — | 47.76 | — | |
| HippoRAGKG Constructor=AutoGraph-R1 (Qwen-3B), Retrieval Paradigm=Graph-based Text2025.10 | 28.11 | — | 47.65 | — | |
| HippoRAG2KG Constructor=AutoGraph-R1 (Qwen-3B), Retrieval Paradigm=Graph-based Text2025.10 | 27.93 | — | 49.13 | — | |
| HippoRAGKG Constructor=DeepSeek-V3, Retrieval Paradigm=Graph-based Text2025.10 | 27.79 | — | 47.23 | — | |
| HippoRAGKG Constructor=Qwen3-235B, Retrieval Paradigm=Graph-based Text2025.10 | 27.72 | — | 47.58 | — | |
| HippoRAGKG Constructor=gemini-2.5-flash-lite, Retrieval Paradigm=Graph-based Text2025.10 | 27.32 | — | 46.36 | — | |
| HippoRAG2KG Constructor=AutoGraph-R1 (Qwen-7B), Retrieval Paradigm=Graph-based Text2025.10 | 27.18 | — | 49.23 | — | |
| HippoRAG2KG Constructor=gpt-oss-120b, Retrieval Paradigm=Graph-based Text2025.10 | 27.03 | — | 46.97 | — | |
| HippoRAG2KG Constructor=Llama-3.3-70B, Retrieval Paradigm=Graph-based Text2025.10 | 26.97 | — | 48.27 | — | |
| HippoRAGKG Constructor=AutoGraph-R1 (Qwen-7B), Retrieval Paradigm=Graph-based Text2025.10 | 26.97 | — | 48.44 | — | |
| HippoRAG2KG Constructor=Qwen2.5-72B-Instruct, Retrieval Paradigm=Graph-based Text2025.10 | 26.71 | — | 48.63 | — | |
| HippoRAGKG Constructor=Qwen2.5-72B-Instruct, Retrieval Paradigm=Graph-based Text2025.10 | 26.42 | — | 46.93 | — | |
| HippoRAGKG Constructor=gpt-oss-120b, Retrieval Paradigm=Graph-based Text2025.10 | 25.37 | — | 45.43 | — | |
| ToG RetrieverKG Constructor=Qwen3-235B, Retrieval Paradigm=Graph Knowledge2025.10 | 21.77 | — | — | — | |
| Triples RetrieverKG Constructor=AutoGraph-R1 (Qwen-3B), Retrieval Paradigm=Graph Knowledge2025.10 | 21.41 | — | — | — | |
| Triples RetrieverKG Constructor=Qwen3-235B, Retrieval Paradigm=Graph Knowledge2025.10 | 20.71 | — | — | — | |
| Triples RetrieverKG Constructor=AutoGraph-R1 (Qwen-7B), Retrieval Paradigm=Graph Knowledge2025.10 | 20.56 | — | — | — | |
| Triples RetrieverKG Constructor=Qwen2.5-72B-Instruct, Retrieval Paradigm=Graph Knowledge2025.10 | 20.07 | — | — | — | |
| Triples RetrieverKG Constructor=gemini-2.5-flash-lite, Retrieval Paradigm=Graph Knowledge2025.10 | 19.91 | — | — | — | |
| ToG RetrieverKG Constructor=gemini-2.5-flash-lite, Retrieval Paradigm=Graph Knowledge2025.10 | 19.6 | — | — | — | |
| Triples RetrieverKG Constructor=Llama-3.3-70B, Retrieval Paradigm=Graph Knowledge2025.10 | 19.55 | — | — | — | |
| ToG RetrieverKG Constructor=AutoGraph-R1 (Qwen-7B), Retrieval Paradigm=Graph Knowledge2025.10 | 19.31 | — | — | — | |
| Triples RetrieverKG Constructor=DeepSeek-V3, Retrieval Paradigm=Graph Knowledge2025.10 | 19.13 | — | — | — | |
| Triples RetrieverKG Constructor=gpt-oss-120b, Retrieval Paradigm=Graph Knowledge2025.10 | 19.1 | — | — | — | |
| ToG RetrieverKG Constructor=GPT-5-mini, Retrieval Paradigm=Graph Knowledge2025.10 | 18.78 | — | — | — | |
| ToG RetrieverKG Constructor=Llama-3.3-70B, Retrieval Paradigm=Graph Knowledge2025.10 | 18.59 | — | — | — | |
| ToG RetrieverKG Constructor=AutoGraph-R1 (Qwen-3B), Retrieval Paradigm=Graph Knowledge2025.10 | 18.42 | — | — | — | |
| ToG RetrieverKG Constructor=Qwen2.5-72B-Instruct, Retrieval Paradigm=Graph Knowledge2025.10 | 18.37 | — | — | — | |
| ToG RetrieverKG Constructor=gpt-oss-120b, Retrieval Paradigm=Graph Knowledge2025.10 | 18.3 | — | — | — | |
| Triples RetrieverKG Constructor=GPT-5-mini, Retrieval Paradigm=Graph Knowledge2025.10 | 18.29 | — | — | — | |
| ToG RetrieverKG Constructor=DeepSeek-V3, Retrieval Paradigm=Graph Knowledge2025.10 | 17.99 | — | — | — | |
| Subgraph RetrieverKG Constructor=Llama-3.3-70B, Retrieval Paradigm=Graph Knowledge2025.10 | 16.11 | — | — | — | |
| Subgraph RetrieverKG Constructor=Qwen3-235B, Retrieval Paradigm=Graph Knowledge2025.10 | 15.83 | — | — | — | |
| Subgraph RetrieverKG Constructor=AutoGraph-R1 (Qwen-7B), Retrieval Paradigm=Graph Knowledge2025.10 | 15.65 | — | — | — | |
| Subgraph RetrieverKG Constructor=gemini-2.5-flash-lite, Retrieval Paradigm=Graph Knowledge2025.10 | 15.18 | — | — | — | |
| Subgraph RetrieverKG Constructor=AutoGraph-R1 (Qwen-3B), Retrieval Paradigm=Graph Knowledge2025.10 | 15.13 | — | — | — | |
| Subgraph RetrieverKG Constructor=Qwen2.5-72B-Instruct, Retrieval Paradigm=Graph Knowledge2025.10 | 14.85 | — | — | — | |
| Subgraph RetrieverKG Constructor=GPT-5-mini, Retrieval Paradigm=Graph Knowledge2025.10 | 14.34 | — | — | — | |
| Subgraph RetrieverKG Constructor=gpt-oss-120b, Retrieval Paradigm=Graph Knowledge2025.10 | 13.39 | — | — | — | |
| Subgraph RetrieverKG Constructor=DeepSeek-V3, Retrieval Paradigm=Graph Knowledge2025.10 | 13.29 | — | — | — | |
| GroupRank-32B*method_category=Reranking Methods, rerank_top_k=50, initial_retriever=Qwen3-Embedding-8B, run_type=single run evaluation2026.02 | — | 55.49 | 65.08 | 73.07 | |
| HippoRAG-v1method_category=Reranking Methods, rerank_top_k=50, initial_retriever=Qwen3-Embedding-8B2026.02 | — | — | 53.2 | — | |
| HippoRAG-v2method_category=Reranking Methods, rerank_top_k=50, initial_retriever=Qwen3-Embedding-8B2026.02 | — | — | 74.7 | — | |
| QRHeads-4B (out-of-box)method_category=Reranking Methods, rerank_top_k=50, initial_retriever=Qwen3-Embedding-8B, status=out-of-box2026.02 | — | 63.12 | 71.22 | 78.99 | |
| QRRanker-4Bmethod_category=Reranking Methods, rerank_top_k=50, initial_retriever=Qwen3-Embedding-8B2026.02 | — | 70.19 | 77.37 | 82.13 | |
| Qwen-Reranker-4B (out-of-box)method_category=Reranking Methods, rerank_top_k=50, initial_retriever=Qwen3-Embedding-8B, status=out-of-box2026.02 | — | 57.6 | 66.37 | 74.26 | |
| Qwen-Reranker-4B (trained)method_category=Reranking Methods, rerank_top_k=50, initial_retriever=Qwen3-Embedding-8B, status=trained2026.02 | — | 61.6 | 69.71 | 77.49 | |
| Qwen3-Embedding-4Bmethod_category=Embedding Methods2026.02 | — | 51.56 | 59.83 | 69.88 | |
| Qwen3-Embedding-8Bmethod_category=Embedding Methods2026.02 | — | 54.35 | 62.55 | 72.47 | |
| SFT-Embedding-8Bmethod_category=Embedding Methods2026.02 | — | 45.11 | 52.93 | 62.03 |