# Multi-hop Question Answering on HotpotQA (SR) Source: [Multi-hop Question Answering on HotpotQA (SR)]() Collection index: [llms.txt]() Data updated: 2026-05-18T06:38:29.194Z Evaluation of model performance on the HotpotQA dataset for multi-hop question answering, specifically reporting the Success Rate (SR). 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 | Multi-hop Question Answering | | Dataset (including any stated split) | HotpotQA | | Metric | Success Rate (SR) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | 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 - [Success Rate (SR)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Success Rate (SR) | Result date | Report paper | | --- | --- | --- | --- | | Gemini-3-Flash | 52 | Not recorded | 6a0a9d04dbc836b87d82b322 | | Gemini-2.5-flash | 47 | Not recorded | 6a0a9d04dbc836b87d82b322 | | MATPO | 46 | Not recorded | 6a0a9d04dbc836b87d82b322 | | Critique-GRPO | 46 | Not recorded | 6a0a9d04dbc836b87d82b322 | | MATPO | 45 | Not recorded | 6a0a9d04dbc836b87d82b322 | | ScalingInter | 44 | Not recorded | 6a0a9d04dbc836b87d82b322 | | ICRL | 44 | Not recorded | 6a0a9d04dbc836b87d82b322 | | GSPO | 44 | Not recorded | 6a0a9d04dbc836b87d82b322 | | ICRL | 44 | Not recorded | 6a0a9d04dbc836b87d82b322 | | Critique-GRPO | 43.5 | Not recorded | 6a0a9d04dbc836b87d82b322 | | GSPO | 42 | Not recorded | 6a0a9d04dbc836b87d82b322 | | ScalingInter | 41.5 | Not recorded | 6a0a9d04dbc836b87d82b322 | | Qwen3-30B-A3B | 40 | Not recorded | 6a0a9d04dbc836b87d82b322 | | GRPO | 38 | Not recorded | 6a0a9d04dbc836b87d82b322 | | GRPO | 38 | Not recorded | 6a0a9d04dbc836b87d82b322 | | Qwen3-8B | 32 | Not recorded | 6a0a9d04dbc836b87d82b322 | | Qwen3-4B | 24 | Not recorded | 6a0a9d04dbc836b87d82b322 | ### Gemini-3-Flash Success Rate (SR): 52. Result ID: 6a0ab3d347f34ce924f03272. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Prompting"}. This record does not assert independent reproduction. ### Gemini-2.5-flash Success Rate (SR): 47. Result ID: 6a0ab3d347f34ce924f03271. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Protocol":"Prompting"}. This record does not assert independent reproduction. ### MATPO Success Rate (SR): 46. Result ID: 6a0ab3d347f34ce924f03276. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B"}. This record does not assert independent reproduction. ### Critique-GRPO Success Rate (SR): 46. Result ID: 6a0ab3d347f34ce924f0327d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction. ### MATPO Success Rate (SR): 45. Result ID: 6a0ab3d347f34ce924f0327c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction. ### ScalingInter Success Rate (SR): 44. Result ID: 6a0ab3d347f34ce924f03275. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B"}. This record does not assert independent reproduction. ### ICRL Success Rate (SR): 44. Result ID: 6a0ab3d347f34ce924f03278. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B"}. This record does not assert independent reproduction. ### GSPO Success Rate (SR): 44. Result ID: 6a0ab3d347f34ce924f0327a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction. ### ICRL Success Rate (SR): 44. Result ID: 6a0ab3d347f34ce924f0327e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction. ### Critique-GRPO Success Rate (SR): 43.5. Result ID: 6a0ab3d347f34ce924f03277. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B"}. This record does not assert independent reproduction. ### GSPO Success Rate (SR): 42. Result ID: 6a0ab3d347f34ce924f03274. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B"}. This record does not assert independent reproduction. ### ScalingInter Success Rate (SR): 41.5. Result ID: 6a0ab3d347f34ce924f0327b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction. ### Qwen3-30B-A3B Success Rate (SR): 40. Result ID: 6a0ab3d347f34ce924f03270. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Protocol":"Prompting"}. This record does not assert independent reproduction. ### GRPO Success Rate (SR): 38. Result ID: 6a0ab3d347f34ce924f03273. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B"}. This record does not assert independent reproduction. ### GRPO Success Rate (SR): 38. Result ID: 6a0ab3d347f34ce924f03279. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction. ### Qwen3-8B Success Rate (SR): 32. Result ID: 6a0ab3d347f34ce924f0326f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Protocol":"Prompting"}. This record does not assert independent reproduction. ### Qwen3-4B Success Rate (SR): 24. Result ID: 6a0ab3d347f34ce924f0326e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Protocol":"Prompting"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.