# Knowledge Base Question Answering on CWQ (Hits@1 and F1) Source: [Knowledge Base Question Answering on CWQ (Hits@1 and F1)]() Collection index: [llms.txt]() Data updated: 2026-06-02T07:52:47.119Z Evaluation of Knowledge Base Question Answering performance on the Complex WebQuestions (CWQ) dataset, reporting Hits@1 and F1 scores. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Knowledge Base Question Answering | | Dataset (including any stated split) | CWQ | | Metric | Hits@1 | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 31 of 31 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Hits@1](): Higher is better; unit: PERCENT. - [F1 Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Hits@1 | Result date | Report paper | | --- | --- | --- | --- | | STAR | 86.3 | Not recorded | 6a0d400ff984739ed6737785 | | RD-P | 83.8 | Not recorded | 6a0d400ff984739ed6737785 | | RoG | 83.3 | Not recorded | 6a0d400ff984739ed6737785 | | Logits-to-Logic | 83.1 | Not recorded | 691414752828b7cc6937e9ae | | Logits-to-Logic | 80.9 | Not recorded | 691414752828b7cc6937e9ae | | DeSQ | 78 | Not recorded | 6a1e636edbd7150506f850df | | ChatKBQA | 76 | Not recorded | 6a1e636edbd7150506f850df | | GCR | 75.8 | Not recorded | 691414752828b7cc6937e9ae | | GCR | 72.7 | Not recorded | 691414752828b7cc6937e9ae | | RoG | 68.3 | Not recorded | 691414752828b7cc6937e9ae | | Graph CoT | 66.1 | Not recorded | 6a0d400ff984739ed6737785 | | RoG | 62.6 | Not recorded | 691414752828b7cc6937e9ae | | GNN-RAG | 61.7 | Not recorded | 691414752828b7cc6937e9ae | | KD-CoT | 55.7 | Not recorded | 691414752828b7cc6937e9ae | | DeepSeek-R1-70B | 55.3 | Not recorded | 6a0d400ff984739ed6737785 | | ToG | 52.1 | Not recorded | 6a0d400ff984739ed6737785 | | ChatGPT+CoT | 47.5 | Not recorded | 691414752828b7cc6937e9ae | | SubgraphRAG | 47.5 | Not recorded | 691414752828b7cc6937e9ae | | DALK | 45.8 | Not recorded | 6a0d400ff984739ed6737785 | | GRAG | 45.6 | Not recorded | 6a0d400ff984739ed6737785 | | Interactive-KBQA | 45.6 | Not recorded | 691414752828b7cc6937e9ae | | Interactive-KBQA | 44 | Not recorded | 691414752828b7cc6937e9ae | | G-Retriever | 42.1 | Not recorded | 6a0d400ff984739ed6737785 | | Llama3.3-70B | 39.5 | Not recorded | 6a0d400ff984739ed6737785 | | ChatGPT+Few-shot | 38.5 | Not recorded | 691414752828b7cc6937e9ae | | ChatGPT | 34.7 | Not recorded | 691414752828b7cc6937e9ae | | Qwen2.5-72B | 32.4 | Not recorded | 6a0d400ff984739ed6737785 | | LLaMA2-7b | 28.4 | Not recorded | 691414752828b7cc6937e9ae | | LLaMA3.1-8b | 28.1 | Not recorded | 691414752828b7cc6937e9ae | | Qwen2-7b | 25.3 | Not recorded | 691414752828b7cc6937e9ae | ### STAR Hits@1: 86.3. Result ID: 6a0d5792fbc373592c04c678. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"Lightweight Retriever","RT":"1.2"}. This record does not assert independent reproduction. ### RD-P Hits@1: 83.8. Result ID: 6a0d5792fbc373592c04c676. [Reporting paper]() Recorded evaluation configuration: {"Setting":"Lightweight Retriever","RT":"1.1"}. This record does not assert independent reproduction. ### RoG Hits@1: 83.3. Result ID: 6a0d5792fbc373592c04c671. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"LLM-centric Retriever","RT":"1.2"}. This record does not assert independent reproduction. ### Logits-to-Logic Hits@1: 83.1. Result ID: 6a1e8bc6fbc373592c1873cf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale Category":"13b model","Backbone":"LLaMA3-13b"}. This record does not assert independent reproduction. ### Logits-to-Logic Hits@1: 80.9. Result ID: 6a1e8bc6fbc373592c1873cc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale Category":"7-8b model","Backbone":"LLaMA3-8b"}. This record does not assert independent reproduction. ### DeSQ Hits@1: 78. Result ID: 6a1e84acfbc373592c183947. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Phi-4"}. This record does not assert independent reproduction. ### ChatKBQA Hits@1: 76. Result ID: 6a1e84acfbc373592c183946. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GCR Hits@1: 75.8. Result ID: 6a1e8bc6fbc373592c1873c2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Hybrid System","Backbone":"LLaMA3-8b+GPT4o-mini"}. This record does not assert independent reproduction. ### GCR Hits@1: 72.7. Result ID: 6a1e8bc6fbc373592c1873c1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Hybrid System","Backbone":"LLaMA3-8b+ChatGPT"}. This record does not assert independent reproduction. ### RoG Hits@1: 68.3. Result ID: 6a1e8bc6fbc373592c1873ce. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale Category":"13b model","Backbone":"LLaMA3-13b"}. This record does not assert independent reproduction. ### Graph CoT Hits@1: 66.1. Result ID: 6a0d5792fbc373592c04c672. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Setting":"LLM-centric Retriever","RT":"513.6"}. This record does not assert independent reproduction. ### RoG Hits@1: 62.6. Result ID: 6a1e8bc6fbc373592c1873ca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale Category":"7-8b model","Backbone":"LLaMA2-Chat-7B"}. This record does not assert independent reproduction. ### GNN-RAG Hits@1: 61.7. Result ID: 6a1e8bc6fbc373592c1873c3. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Hybrid System","Backbone":"LLaMA2-7b+GNN"}. This record does not assert independent reproduction. ### KD-CoT Hits@1: 55.7. Result ID: 6a1e8bc6fbc373592c1873c9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale Category":"7-8b model","Backbone":"LLaMA2-7b"}. This record does not assert independent reproduction. ### DeepSeek-R1-70B Hits@1: 55.3. Result ID: 6a0d5792fbc373592c04c670. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"LLM without KG","RT":"17.2"}. This record does not assert independent reproduction. ### ToG Hits@1: 52.1. Result ID: 6a0d5792fbc373592c04c673. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"LLM-centric Retriever","RT":"90.4"}. This record does not assert independent reproduction. ### ChatGPT+CoT Hits@1: 47.5. Result ID: 6a1e8bc6fbc373592c1873c0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Closed-source business model"}. This record does not assert independent reproduction. ### SubgraphRAG Hits@1: 47.5. Result ID: 6a1e8bc6fbc373592c1873cb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale Category":"7-8b model","Backbone":"Llama3.1-8B"}. This record does not assert independent reproduction. ### DALK Hits@1: 45.8. Result ID: 6a0d5792fbc373592c04c674. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"Lightweight Retriever","RT":"72.6"}. This record does not assert independent reproduction. ### GRAG Hits@1: 45.6. Result ID: 6a0d5792fbc373592c04c677. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"Lightweight Retriever","RT":"7.2"}. This record does not assert independent reproduction. ### Interactive-KBQA Hits@1: 45.6. Result ID: 6a1e8bc6fbc373592c1873cd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale Category":"13b model","Backbone":"LLaMA2-13b"}. This record does not assert independent reproduction. ### Interactive-KBQA Hits@1: 44. Result ID: 6a1e8bc6fbc373592c1873c8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale Category":"7-8b model","Backbone":"Mistral-7B"}. This record does not assert independent reproduction. ### G-Retriever Hits@1: 42.1. Result ID: 6a0d5792fbc373592c04c675. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Setting":"Lightweight Retriever","RT":"6.7"}. This record does not assert independent reproduction. ### Llama3.3-70B Hits@1: 39.5. Result ID: 6a0d5792fbc373592c04c66f. [Reporting paper]() Recorded evaluation configuration: {"Setting":"LLM without KG","RT":"3.6"}. This record does not assert independent reproduction. ### ChatGPT+Few-shot Hits@1: 38.5. Result ID: 6a1e8bc6fbc373592c1873bf. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Closed-source business model"}. This record does not assert independent reproduction. ### ChatGPT Hits@1: 34.7. Result ID: 6a1e8bc6fbc373592c1873be. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Closed-source business model"}. This record does not assert independent reproduction. ### Qwen2.5-72B Hits@1: 32.4. Result ID: 6a0d5792fbc373592c04c66e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"LLM without KG","RT":"4.3"}. This record does not assert independent reproduction. ### LLaMA2-7b Hits@1: 28.4. Result ID: 6a1e8bc6fbc373592c1873c6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Scale Category":"7-8b model"}. This record does not assert independent reproduction. ### LLaMA3.1-8b Hits@1: 28.1. Result ID: 6a1e8bc6fbc373592c1873c7. [Reporting paper]() Recorded evaluation configuration: {"Model Scale Category":"7-8b model"}. This record does not assert independent reproduction. ### Qwen2-7b Hits@1: 25.3. Result ID: 6a1e8bc6fbc373592c1873c5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale Category":"7-8b model"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.