# Multi-Hop Question Answering on MuSiQue (Exact Match %) Source: [Multi-Hop Question Answering on MuSiQue (Exact Match %)]() Collection index: [llms.txt]() Data updated: 2026-07-08T07:46:57.965Z Evaluation of model performance on the MuSiQue dataset for Multi-Hop Question Answering, reporting the Exact Match percentage. 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) | MuSiQue | | Metric | Exact Match (%) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | 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 - [Exact Match (%)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Exact Match (%) | Result date | Report paper | | --- | --- | --- | --- | | StepSearch | 22.6 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | IGRPO | 20.2 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | GiGPO | 18.9 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | ZeroSearch | 18.4 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | Tree-GRPO | 17.5 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | StepSearch | 17.4 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | IGPO | 16.8 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | Search-R1 | 14.6 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | IGRPO | 14.1 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | GiGPO | 12.6 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | IGPO | 12.3 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | Tree-GRPO | 11.9 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | Search-R1 | 10.3 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | ZeroSearch | 9.8 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | R1-Instruct | 7.2 | Not recorded | 6a4dd9cd4da6617fe9198d1d | | R1-Instruct | 6 | Not recorded | 6a4dd9cd4da6617fe9198d1d | ### StepSearch Exact Match (%): 22.6. Result ID: 6a4e006260b88bb339d76ef9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Type":"Chain-based"}. This record does not assert independent reproduction. ### IGRPO Exact Match (%): 20.2. Result ID: 6a4e006260b88bb339d76efd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Type":"Tree-based"}. This record does not assert independent reproduction. ### GiGPO Exact Match (%): 18.9. Result ID: 6a4e006260b88bb339d76efa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Type":"Chain-based"}. This record does not assert independent reproduction. ### ZeroSearch Exact Match (%): 18.4. Result ID: 6a4e006260b88bb339d76ef8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Type":"Chain-based"}. This record does not assert independent reproduction. ### Tree-GRPO Exact Match (%): 17.5. Result ID: 6a4e006260b88bb339d76efc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Type":"Tree-based"}. This record does not assert independent reproduction. ### StepSearch Exact Match (%): 17.4. Result ID: 6a4e006260b88bb339d76ef1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Type":"Chain-based"}. This record does not assert independent reproduction. ### IGPO Exact Match (%): 16.8. Result ID: 6a4e006260b88bb339d76efb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Type":"Chain-based"}. This record does not assert independent reproduction. ### Search-R1 Exact Match (%): 14.6. Result ID: 6a4e006260b88bb339d76ef7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Type":"Chain-based"}. This record does not assert independent reproduction. ### IGRPO Exact Match (%): 14.1. Result ID: 6a4e006260b88bb339d76ef5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Type":"Tree-based"}. This record does not assert independent reproduction. ### GiGPO Exact Match (%): 12.6. Result ID: 6a4e006260b88bb339d76ef2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Type":"Chain-based"}. This record does not assert independent reproduction. ### IGPO Exact Match (%): 12.3. Result ID: 6a4e006260b88bb339d76ef3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Type":"Chain-based"}. This record does not assert independent reproduction. ### Tree-GRPO Exact Match (%): 11.9. Result ID: 6a4e006260b88bb339d76ef4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Type":"Tree-based"}. This record does not assert independent reproduction. ### Search-R1 Exact Match (%): 10.3. Result ID: 6a4e006260b88bb339d76eef. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Type":"Chain-based"}. This record does not assert independent reproduction. ### ZeroSearch Exact Match (%): 9.8. Result ID: 6a4e006260b88bb339d76ef0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Type":"Chain-based"}. This record does not assert independent reproduction. ### R1-Instruct Exact Match (%): 7.2. Result ID: 6a4e006260b88bb339d76eee. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-3B-Instruct","Type":"RL Training"}. This record does not assert independent reproduction. ### R1-Instruct Exact Match (%): 6. Result ID: 6a4e006260b88bb339d76ef6. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Type":"RL Training"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.