Multi-hop Question Answering on HotpotQA (test)
80.79F1Gemini-2.5-Flash
Evaluation Results
| Method | Links | ||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Gemini-2.5-FlashType=API2026.02 | 80.79 | — | — | — | — | — | — | 65.68 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GPT-5-miniType=API2026.02 | 80.19 | — | — | — | — | — | — | 63.63 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Deepseek-V3.2Type=API2026.02 | 79.91 | — | — | — | — | — | — | 64.15 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-30BType=Pre-Trained2026.02 | 78.33 | — | — | — | — | — | — | 63.05 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CRAFT7BType=Trained2026.02 | 78 | — | — | — | — | — | — | 63.25 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DenoiseFlowBackbone LLM=GPT-4o-mini, System Category=Autonomous Multi-agent2026.02 | 77.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| JudgeFlowBackbone LLM=GPT-4o-mini, System Category=Autonomous Multi-agent2026.02 | 77.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MermaidFlowBackbone LLM=GPT-4o-mini, System Category=Autonomous Multi-agent2026.02 | 77.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HELPBackbone=Llama-3.3-70B-Instruct2026.02 | 75.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HippoRAG 2Generator=Llama3.3-70B-Instruct2026.02 | 75.5 | — | — | — | — | — | — | 62.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HippoRAG2Backbone=Llama-3.3-70B-Instruct2026.02 | 75.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-4BType=Pre-Trained2026.02 | 75.32 | — | — | — | — | — | — | 60.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NV-Embed-v2 (7B)Generator=Llama3.3-70B-Instruct2026.02 | 75.3 | — | — | — | — | — | — | 62.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NV-Embed-v2Backbone=Llama-3.3-70B-Instruct2026.02 | 75.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MaASBackbone LLM=GPT-4o-mini, System Category=Autonomous Multi-agent2026.02 | 75.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DyLANBackbone LLM=GPT-4o-mini, System Category=Hand-crafted Multi-agent2026.02 | 74.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AFlowBackbone LLM=GPT-4o-mini, System Category=Autonomous Multi-agent2026.02 | 73.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GritLM-7BBackbone=Llama-3.3-70B-Instruct2026.02 | 73.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GPTSwarmBackbone LLM=GPT-4o-mini, System Category=Autonomous Multi-agent2026.02 | 73.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HyperGraphRAGBackbone=Llama-3.3-70B-Instruct, Local Reproduction=true2026.02 | 73.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MultiCube-RAGGenerator=Llama3.3-70B-Instruct2026.02 | 72.9 | — | — | — | — | — | — | 61.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LLM-BlenderBackbone LLM=GPT-4o-mini, System Category=Hand-crafted Multi-agent2026.02 | 72.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MultiCube-RAGGenerator=GPT-4o-mini2026.02 | 71.5 | — | — | — | — | — | — | 57.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CatRAGQA reader=Llama-3.3-70B-Instruct2026.02 | 71.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HippoRAG 2Generator=GPT-4o-mini2026.02 | 71.1 | — | — | — | — | — | — | 56.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NV-Embed-v2 (7B)Generator=GPT-4o-mini2026.02 | 71 | — | — | — | — | — | — | 57.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GTE-Qwen2-7BBackbone=Llama-3.3-70B-Instruct2026.02 | 71 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RelinkMethod Type=Proposed, Backbone LLM=deepseek-v3-03242026.01 | 70.4 | — | — | — | — | — | — | 55.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LLM-DebateBackbone LLM=GPT-4o-mini, System Category=Hand-crafted Multi-agent2026.02 | 70.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RAPTORGenerator=Llama3.3-70B-Instruct2026.02 | 69.5 | — | — | — | — | — | — | 56.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RAPTORBackbone=Llama-3.3-70B-Instruct2026.02 | 69.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OPERA (MAPGRPO)Type=RL2025.08 | 69.5 | — | — | — | — | — | — | 57.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HippoRAG 2QA reader=Llama-3.3-70B-Instruct, Retrieval=Structure-Aware RAG2026.02 | 69.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| IGMiRAG2026.02 | 69.1 | — | — | — | — | — | — | 54 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CoT SCBackbone LLM=GPT-4o-mini, System Category=Single-agent2026.02 | 68.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GraphRAGGenerator=Llama3.3-70B-Instruct2026.02 | 68.6 | — | — | — | — | — | — | 55.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GraphRAGBackbone=Llama-3.3-70B-Instruct2026.02 | 68.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LightRAGQA reader=Llama-3.3-70B-Instruct, Retrieval=Structure-Aware RAG2026.02 | 68.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2.5-7BType=Trained2026.02 | 68.12 | — | — | — | — | — | — | 53.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| IOBackbone LLM=GPT-4o-mini, System Category=Single-agent2026.02 | 68.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AceSearcher-8Bcategory=Advanced RAG2026.04 | 68.07 | — | — | — | — | — | — | 37.19 | 42.91 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2.5-7BType=Pre-Trained2026.02 | 67.92 | — | — | — | — | — | — | 53.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CoTBackbone LLM=GPT-4o-mini, System Category=Single-agent2026.02 | 67.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GraphRAGGenerator=GPT-4o-mini2026.02 | 67.6 | — | — | — | — | — | — | 51.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ProGraph-R1Base Model=Qwen2.5-7B-Instruct, Optimization Method=Training, Knowledge Interaction=Graph-based knowledge2026.01 | 67.57 | — | — | — | — | — | — | 60.94 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CRAFT3BType=Trained2026.02 | 67.1 | — | — | — | — | — | — | 53.15 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RankCoTGenerator=GPT-4o-mini2026.02 | 65.8 | — | — | — | — | — | — | 53.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RoMemMemory category=Structured memory baselines2026.06 | 65.8 | — | — | — | — | — | — | 50.9 | 60.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | 79.9 | — | — | — | — | — | — | — | |
| Naive RAGTop-k retrieved slices=52026.02 | 65.1 | — | — | — | — | — | — | 50.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HippoRAGMethod Type=Hybrid, Backbone LLM=deepseek-v3-03242026.01 | 64.7 | — | — | — | — | — | — | 49.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RAPTORGenerator=GPT-4o-mini2026.02 | 64.7 | — | — | — | — | — | — | 50.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| text-embedding-3-smallQA reader=Llama-3.3-70B-Instruct, Retrieval=Standard Retrieval2026.02 | 64.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ADASBackbone LLM=GPT-4o-mini, System Category=Autonomous Multi-agent2026.02 | 64.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RAPTORQA reader=Llama-3.3-70B-Instruct, Retrieval=Structure-Aware RAG2026.02 | 64.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| IRCoTGenerator=GPT-4o-mini2026.02 | 64.1 | — | — | — | — | — | — | 53.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NodeRAG2026.02 | 64 | — | — | — | — | — | — | 50.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HippoRAGGenerator=Llama3.3-70B-Instruct2026.02 | 63.5 | — | — | — | — | — | — | 52.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HippoRAGBackbone=Llama-3.3-70B-Instruct2026.02 | 63.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FABLE(nodes)selection_strategy=LLM-guided, bi_path_fusion=true, retrieval_unit=nodes2026.01 | 63.43 | — | — | — | — | — | — | 48 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BM25Generator=Llama3.3-70B-Instruct2026.02 | 63.4 | — | — | — | — | — | — | 52 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BM25Backbone=Llama-3.3-70B-Instruct2026.02 | 63.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RAPTORMethod Type=Text-based, Backbone LLM=deepseek-v3-03242026.01 | 63.1 | — | — | — | — | — | — | 47.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HERA (Qwen-3-14B)category=Ours2026.04 | 63.03 | — | — | — | — | — | — | 52.5 | 55.38 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GTR (T5-base)QA reader=Llama-3.3-70B-Instruct, Retrieval=Standard Retrieval2026.02 | 62.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GTRBackbone=Llama-3.3-70B-Instruct2026.02 | 62.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Naive RAGTop-k retrieved slices=32026.02 | 62.7 | — | — | — | — | — | — | 48.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Graph-R1Base Model=Qwen2.5-7B-Instruct, Optimization Method=Training, Knowledge Interaction=Graph-based knowledge2026.01 | 62.69 | — | — | — | — | — | — | 57.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2.5-3BType=Pre-Trained2026.02 | 62.58 | — | — | — | — | — | — | 49.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GenGroundBackbone=GPT-4o, Retrieval=with Retrieval2025.06 | 62.37 | 5 | 60.5 | 63 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ContrieverQA reader=Llama-3.3-70B-Instruct, Retrieval=Standard Retrieval2026.02 | 62.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ContrieverBackbone=Llama-3.3-70B-Instruct2026.02 | 62.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FABLE(docs)selection_strategy=LLM-guided, bi_path_fusion=true, retrieval_unit=documents2026.01 | 62.22 | — | — | — | — | — | — | 46.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DEC (Ours)Backbone=GPT-4o, Retrieval=with Retrieval2025.06 | 62.11 | 3.32 | 58.52 | 60.32 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ProGraph-R1Base Model=Qwen2.5-3B-Instruct, Optimization Method=Training, Knowledge Interaction=Graph-based knowledge2026.01 | 61.9 | — | — | — | — | — | — | 55.47 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Self-RefineBackbone LLM=GPT-4o-mini, System Category=Hand-crafted Multi-agent2026.02 | 60.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| EviOmniModel size=7B, Method Category=RAR (Reasoning-Augmented), Context=Out-of-domain2025.07 | 60.3 | — | — | — | — | — | — | 28.02 | — | — | — | — | — | — | — | 16.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2.5-3BType=Trained2026.02 | 60.23 | — | — | — | — | — | — | 46.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HippoRAGGenerator=GPT-4o-mini2026.02 | 60 | — | — | — | — | — | — | 46.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CoTModel size=7B, Method Category=RAR (Reasoning-Augmented)2025.07 | 59.84 | — | — | — | — | — | — | 27.25 | — | — | — | — | — | — | — | 3.32 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BM25Generator=GPT-4o-mini2026.02 | 59.8 | — | — | — | — | — | — | 49.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Direct CoTBackbone=GPT-4o, Retrieval=without Retrieval2025.06 | 59.31 | 1 | 55 | 58.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| IQATRLLM Backbone=Llama-3.1-8B, Retriever=Contriever, Retriever Fine-tuning=ReSCORE2025.05 | 59.3 | — | — | — | — | — | — | 47.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ARBORBackbone=Qwen3-14B2026.06 | 59.1 | — | — | — | — | — | — | 48.6 | 63.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MetaTuner-IBackbone=Qwen2.5-3B, Method Category=Hybrid2025.09 | 59.05 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CRAFT1.5BType=Trained2026.02 | 59.04 | — | — | — | — | — | — | 45.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LinearRAGBackbone=Llama-3.3-70B-Instruct, Local Reproduction=true2026.02 | 59 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HippoRAG2category=Structure-Enhanced RAG2026.01 | 58.87 | — | — | — | — | — | — | 41 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DeepResearcher-7Bcategory=Advanced RAG2026.04 | 58.86 | — | — | — | — | — | — | 35.23 | 40.65 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MetaTuner-JBackbone=Qwen2.5-3B, Method Category=Hybrid2025.09 | 58.85 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Abstract-CoT (Warm-up + RL)Model=Qwen3-8B, Tokens=1712026.04 | 58.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SEERModel size=7B, Method Category=VAR (Context Compression)2025.07 | 58.63 | — | — | — | — | — | — | 25.81 | — | — | — | — | — | — | — | 14.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ARBORBackbone=Qwen3-8B2026.06 | 58.6 | — | — | — | — | — | — | 47.8 | 62.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FullModel size=7B, Method Category=WR (Vanilla Retrieval)2025.07 | 58.5 | — | — | — | — | — | — | 25.82 | — | — | — | — | — | — | — | 1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OPERA (CoT)Type=CoT2025.08 | 58.5 | — | — | — | — | — | — | 44.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Adaptive-NoteLLM Backbone=GPT-3.5, Retriever=BM252025.05 | 58.4 | — | — | — | — | — | — | 45.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GRPOBackbone=Qwen3-14B2026.06 | 58.4 | — | — | — | — | — | — | 48.1 | 62.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ExSearchcategory=Advanced RAG, Backbone=Qwen2.5-7B2026.04 | 58.22 | — | — | — | — | — | — | 45.11 | 55.04 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FilCoModel size=7B, Method Category=VAR (Context Compression)2025.07 | 58.15 | — | — | — | — | — | — | 25.27 | — | — | — | — | — | — | — | 17.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SearChainBackbone=GPT-4o, Retrieval=with Retrieval2025.06 | 58.12 | 2.24 | 53.8 | 58 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SFT + RLModel=Qwen3-8B, Tokens=7352026.04 | 58.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — |