# Open-domain multi-hop question answering on 2Wiki Source: [Open-domain multi-hop question answering on 2Wiki]() Collection index: [llms.txt]() Data updated: 2026-04-29T06:52:47.269Z Evaluation of model performance on the 2Wiki dataset for open-domain multi-hop question answering, specifically measured using Answer F1, Citation F1, and Joint 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 | Open-domain multi-hop question answering | | Dataset (including any stated split) | 2Wiki | | Metric | Answer F1 | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 - [Answer F1](): Higher is better; unit: PERCENT. - [Citation F1](): Higher is better; unit: PERCENT. - [Joint F1](): Higher is better; unit: PERCENT. ## Recorded results | Method | Answer F1 | Result date | Report paper | | --- | --- | --- | --- | | RAG-RL | 75.06 | Not recorded | 68fa78c51e209131cea7a6d3 | | RAG-RL | 74.31 | Not recorded | 68fa78c51e209131cea7a6d3 | | RAG-RL | 72.61 | Not recorded | 68fa78c51e209131cea7a6d3 | | RAG-RL | 71.82 | Not recorded | 68fa78c51e209131cea7a6d3 | | RAG-RL | 71.04 | Not recorded | 68fa78c51e209131cea7a6d3 | | RAG-RL | 70.77 | Not recorded | 68fa78c51e209131cea7a6d3 | | Qwen2.5-7B-Instruct | 70.29 | Not recorded | 68fa78c51e209131cea7a6d3 | | RAG-RL | 68.92 | Not recorded | 68fa78c51e209131cea7a6d3 | | RAG-RL | 68.56 | Not recorded | 68fa78c51e209131cea7a6d3 | | RAG-RL | 67.99 | Not recorded | 68fa78c51e209131cea7a6d3 | | RAG-RL | 66.59 | Not recorded | 68fa78c51e209131cea7a6d3 | | Qwen2.5-7B-Instruct | 48.71 | Not recorded | 68fa78c51e209131cea7a6d3 | ### RAG-RL Answer F1: 75.06. Result ID: 69f1aaad47f34ce924d79c94. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Eval. Setting":"Ideal Retrieval","Curriculum":"Min-Max"}. This record does not assert independent reproduction. ### RAG-RL Answer F1: 74.31. Result ID: 69f1aaad47f34ce924d79c93. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Eval. Setting":"Ideal Retrieval","Curriculum":"Linear"}. This record does not assert independent reproduction. ### RAG-RL Answer F1: 72.61. Result ID: 69f1aaad47f34ce924d79c95. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Eval. Setting":"Ideal Retrieval","Curriculum":"Base-Answerable"}. This record does not assert independent reproduction. ### RAG-RL Answer F1: 71.82. Result ID: 69f1aaad47f34ce924d79c92. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Eval. Setting":"Ideal Retrieval","Curriculum":"Max"}. This record does not assert independent reproduction. ### RAG-RL Answer F1: 71.04. Result ID: 69f1aaad47f34ce924d79c96. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Eval. Setting":"Ideal Retrieval","Curriculum":"Base-Unanswerable"}. This record does not assert independent reproduction. ### RAG-RL Answer F1: 70.77. Result ID: 69f1aaad47f34ce924d79c8e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Eval. Setting":"Distractor","Curriculum":"Min-Max"}. This record does not assert independent reproduction. ### Qwen2.5-7B-Instruct Answer F1: 70.29. Result ID: 69f1aaad47f34ce924d79c91. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Eval. Setting":"Ideal Retrieval","Curriculum":"Baseline"}. This record does not assert independent reproduction. ### RAG-RL Answer F1: 68.92. Result ID: 69f1aaad47f34ce924d79c8d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Eval. Setting":"Distractor","Curriculum":"Linear"}. This record does not assert independent reproduction. ### RAG-RL Answer F1: 68.56. Result ID: 69f1aaad47f34ce924d79c8f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Eval. Setting":"Distractor","Curriculum":"Base-Answerable"}. This record does not assert independent reproduction. ### RAG-RL Answer F1: 67.99. Result ID: 69f1aaad47f34ce924d79c8c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Eval. Setting":"Distractor","Curriculum":"Max"}. This record does not assert independent reproduction. ### RAG-RL Answer F1: 66.59. Result ID: 69f1aaad47f34ce924d79c90. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Eval. Setting":"Distractor","Curriculum":"Base-Unanswerable"}. This record does not assert independent reproduction. ### Qwen2.5-7B-Instruct Answer F1: 48.71. Result ID: 69f1aaad47f34ce924d79c8b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Eval. Setting":"Distractor","Curriculum":"Baseline"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.