# Multi-hop Question Answering on Bamboogle (EM, RAG Improvement) Source: [Multi-hop Question Answering on Bamboogle (EM, RAG Improvement)]() Collection index: [llms.txt]() Data updated: 2026-05-14T07:58:39.982Z Evaluation of model performance on the Bamboogle dataset for Multi-hop Question Answering, focusing on Exact Match accuracy and improvement relative to a Vanilla RAG baseline. 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) | Bamboogle | | Metric | Exact Match (EM) | | 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 - [Exact Match (EM)](): Higher is better; unit: PERCENT. - [Improvement over Vanilla RAG](): Higher is better; unit: SCALAR. ## Recorded results | Method | Exact Match (EM) | Result date | Report paper | | --- | --- | --- | --- | | PyRAG | 45.5 | Not recorded | 6a0556f0dbc836b87d828b5a | | PyRAG | 41.5 | Not recorded | 6a0556f0dbc836b87d828b5a | | ITER-RETGEN | 33.6 | Not recorded | 6a0556f0dbc836b87d828b5a | | ITER-RETGEN | 29.6 | Not recorded | 6a0556f0dbc836b87d828b5a | | IRCoT | 28.3 | Not recorded | 6a0556f0dbc836b87d828b5a | | Self-Ask | 26.1 | Not recorded | 6a0556f0dbc836b87d828b5a | | IRCoT | 24.3 | Not recorded | 6a0556f0dbc836b87d828b5a | | CoT | 23.2 | Not recorded | 6a0556f0dbc836b87d828b5a | | Self-Ask | 22.1 | Not recorded | 6a0556f0dbc836b87d828b5a | | Vanilla RAG | 21.6 | Not recorded | 6a0556f0dbc836b87d828b5a | | CoT | 19.6 | Not recorded | 6a0556f0dbc836b87d828b5a | | Direct Inference | 17.6 | Not recorded | 6a0556f0dbc836b87d828b5a | | Vanilla RAG | 16 | Not recorded | 6a0556f0dbc836b87d828b5a | | Direct Inference | 12 | Not recorded | 6a0556f0dbc836b87d828b5a | ### PyRAG Exact Match (EM): 45.5. Result ID: 6a05809f47f34ce924ec4d29. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-72B-Instruct"}. This record does not assert independent reproduction. ### PyRAG Exact Match (EM): 41.5. Result ID: 6a05809f47f34ce924ec4d22. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct"}. This record does not assert independent reproduction. ### ITER-RETGEN Exact Match (EM): 33.6. Result ID: 6a05809f47f34ce924ec4d28. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-72B-Instruct"}. This record does not assert independent reproduction. ### ITER-RETGEN Exact Match (EM): 29.6. Result ID: 6a05809f47f34ce924ec4d21. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct"}. This record does not assert independent reproduction. ### IRCoT Exact Match (EM): 28.3. Result ID: 6a05809f47f34ce924ec4d27. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-72B-Instruct"}. This record does not assert independent reproduction. ### Self-Ask Exact Match (EM): 26.1. Result ID: 6a05809f47f34ce924ec4d26. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-72B-Instruct"}. This record does not assert independent reproduction. ### IRCoT Exact Match (EM): 24.3. Result ID: 6a05809f47f34ce924ec4d20. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct"}. This record does not assert independent reproduction. ### CoT Exact Match (EM): 23.2. Result ID: 6a05809f47f34ce924ec4d1d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct"}. This record does not assert independent reproduction. ### Self-Ask Exact Match (EM): 22.1. Result ID: 6a05809f47f34ce924ec4d1f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct"}. This record does not assert independent reproduction. ### Vanilla RAG Exact Match (EM): 21.6. Result ID: 6a05809f47f34ce924ec4d25. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-72B-Instruct"}. This record does not assert independent reproduction. ### CoT Exact Match (EM): 19.6. Result ID: 6a05809f47f34ce924ec4d24. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-72B-Instruct"}. This record does not assert independent reproduction. ### Direct Inference Exact Match (EM): 17.6. Result ID: 6a05809f47f34ce924ec4d23. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-72B-Instruct"}. This record does not assert independent reproduction. ### Vanilla RAG Exact Match (EM): 16. Result ID: 6a05809f47f34ce924ec4d1e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct"}. This record does not assert independent reproduction. ### Direct Inference Exact Match (EM): 12. Result ID: 6a05809f47f34ce924ec4d1c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.