# Multi-hop Question Answering on MuSiQue (% Cost Saving) Source: [Multi-hop Question Answering on MuSiQue (% Cost Saving)]() Collection index: [llms.txt]() Data updated: 2026-05-27T10:33:13.796Z Evaluation of model performance on the MuSiQue dataset for Multi-hop Question Answering, specifically reporting the percentage cost saving achieved. 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 | Cost Saving | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 14 of 14 | | 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 - [Cost Saving](): Higher is better; unit: PERCENT. ## Recorded results | Method | Cost Saving | Result date | Report paper | | --- | --- | --- | --- | | Adaptive-RAG (Extended) | 159.2 | Not recorded | 6a167adbdbd7150506f80331 | | Adaptive-RAG (Extended) | 159.2 | Not recorded | 6a167adbdbd7150506f80331 | | CARROT-RoBERTa (LLM-only) | 92.3 | Not recorded | 6a167adbdbd7150506f80331 | | BRANE | 89.4 | Not recorded | 6a167adbdbd7150506f80331 | | CARROT-KNN (Extended) | 77 | Not recorded | 6a167adbdbd7150506f80331 | | CARROT-RoBERTa (Extended) | 59.2 | Not recorded | 6a167adbdbd7150506f80331 | | CARROT-KNN (Extended) | 57.5 | Not recorded | 6a167adbdbd7150506f80331 | | CARROT-RoBERTa (Extended) | 44.6 | Not recorded | 6a167adbdbd7150506f80331 | | BRANE | 40.7 | Not recorded | 6a167adbdbd7150506f80331 | | BRANE | 17.3 | Not recorded | 6a167adbdbd7150506f80331 | | CARROT-RoBERTa (Extended) | 2.4 | Not recorded | 6a167adbdbd7150506f80331 | | Most Accurate Static (Murakkab) | 0 | Not recorded | 6a167adbdbd7150506f80331 | | Most Accurate Static (Murakkab) | 0 | Not recorded | 6a167adbdbd7150506f80331 | | Most Accurate Static (Murakkab) | 0 | Not recorded | 6a167adbdbd7150506f80331 | ### Adaptive-RAG (Extended) Cost Saving: 159.2. Result ID: 6a16c853fbc373592c10a047. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"95%"}. This record does not assert independent reproduction. ### Adaptive-RAG (Extended) Cost Saving: 159.2. Result ID: 6a16c853fbc373592c10a04c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"90%"}. This record does not assert independent reproduction. ### CARROT-RoBERTa (LLM-only) Cost Saving: 92.3. Result ID: 6a16c853fbc373592c10a04d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"90%"}. This record does not assert independent reproduction. ### BRANE Cost Saving: 89.4. Result ID: 6a16c853fbc373592c10a045. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"100%"}. This record does not assert independent reproduction. ### CARROT-KNN (Extended) Cost Saving: 77. Result ID: 6a16c853fbc373592c10a048. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"95%"}. This record does not assert independent reproduction. ### CARROT-RoBERTa (Extended) Cost Saving: 59.2. Result ID: 6a16c853fbc373592c10a049. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"95%"}. This record does not assert independent reproduction. ### CARROT-KNN (Extended) Cost Saving: 57.5. Result ID: 6a16c853fbc373592c10a04e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"90%"}. This record does not assert independent reproduction. ### CARROT-RoBERTa (Extended) Cost Saving: 44.6. Result ID: 6a16c853fbc373592c10a04f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"90%"}. This record does not assert independent reproduction. ### BRANE Cost Saving: 40.7. Result ID: 6a16c853fbc373592c10a050. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"90%"}. This record does not assert independent reproduction. ### BRANE Cost Saving: 17.3. Result ID: 6a16c853fbc373592c10a04a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"95%"}. This record does not assert independent reproduction. ### CARROT-RoBERTa (Extended) Cost Saving: 2.4. Result ID: 6a16c853fbc373592c10a044. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"100%"}. This record does not assert independent reproduction. ### Most Accurate Static (Murakkab) Cost Saving: 0. Result ID: 6a16c853fbc373592c10a043. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"100%"}. This record does not assert independent reproduction. ### Most Accurate Static (Murakkab) Cost Saving: 0. Result ID: 6a16c853fbc373592c10a046. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"95%"}. This record does not assert independent reproduction. ### Most Accurate Static (Murakkab) Cost Saving: 0. Result ID: 6a16c853fbc373592c10a04b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Accuracy Goal":"90%"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.