# Question Answering on Natural Questions (NQ) (EM and Performance Gain Metrics) Source: [Question Answering on Natural Questions (NQ) (EM and Performance Gain Metrics)]() Collection index: [llms.txt]() Data updated: 2026-04-09T23:58:42.336Z Evaluation of model performance on the Natural Questions (NQ) dataset for Question Answering, focusing on Exact Match (EM) score and relative Performance Gain. 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 | Question Answering | | Dataset (including any stated split) | Natural Questions (NQ) | | Metric | Exact Match (EM) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | 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. - [Performance Gain](): Higher is better; unit: SCALAR. ## Recorded results | Method | Exact Match (EM) | Result date | Report paper | | --- | --- | --- | --- | | PatchRAG | 49.8 | Not recorded | 69d732af7d0172140ec76c58 | | ChatQA-1.5 | 42.4 | Not recorded | 69d732af7d0172140ec76c58 | | RAFT | 41.9 | Not recorded | 69d732af7d0172140ec76c58 | | Golden-only | 40.2 | Not recorded | 69d732af7d0172140ec76c58 | | Auto-RAG | 37.9 | Not recorded | 69d732af7d0172140ec76c58 | | Self-RAG | 36.4 | Not recorded | 69d732af7d0172140ec76c58 | | RAFT | 36.4 | Not recorded | 69d732af7d0172140ec76c58 | | PatchRAG | 36.2 | Not recorded | 69d732af7d0172140ec76c58 | | Golden-only | 35.3 | Not recorded | 69d732af7d0172140ec76c58 | | Standard RAG | 28.7 | Not recorded | 69d732af7d0172140ec76c58 | ### PatchRAG Exact Match (EM): 49.8. Result ID: 69d83d1e47f34ce924c028df. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Step":"t","Generator":"Llama-3 8B","Retriever":"bge-m3"}. This record does not assert independent reproduction. ### ChatQA-1.5 Exact Match (EM): 42.4. Result ID: 69d83d1e47f34ce924c028d9. [Reporting paper]() Recorded evaluation configuration: {"Step":"-","Generator":"Llama-3 8B","Retriever":"bge-m3"}. This record does not assert independent reproduction. ### RAFT Exact Match (EM): 41.9. Result ID: 69d83d1e47f34ce924c028de. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Step":"t","Generator":"Llama-3 8B","Retriever":"bge-m3"}. This record does not assert independent reproduction. ### Golden-only Exact Match (EM): 40.2. Result ID: 69d83d1e47f34ce924c028dd. [Reporting paper]() Recorded evaluation configuration: {"Step":"t","Generator":"Llama-3 8B","Retriever":"bge-m3"}. This record does not assert independent reproduction. ### Auto-RAG Exact Match (EM): 37.9. Result ID: 69d83d1e47f34ce924c028d8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Step":"-","Generator":"Llama-3 8B","Retriever":"bge-m3"}. This record does not assert independent reproduction. ### Self-RAG Exact Match (EM): 36.4. Result ID: 69d83d1e47f34ce924c028d7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Step":"-","Generator":"Llama-3 8B","Retriever":"bge-m3"}. This record does not assert independent reproduction. ### RAFT Exact Match (EM): 36.4. Result ID: 69d83d1e47f34ce924c028db. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Step":"\< t","Generator":"Llama-3 8B","Retriever":"bge-m3"}. This record does not assert independent reproduction. ### PatchRAG Exact Match (EM): 36.2. Result ID: 69d83d1e47f34ce924c028dc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Step":"\< t","Generator":"Llama-3 8B","Retriever":"bge-m3"}. This record does not assert independent reproduction. ### Golden-only Exact Match (EM): 35.3. Result ID: 69d83d1e47f34ce924c028da. [Reporting paper]() Recorded evaluation configuration: {"Step":"\< t","Generator":"Llama-3 8B","Retriever":"bge-m3"}. This record does not assert independent reproduction. ### Standard RAG Exact Match (EM): 28.7. Result ID: 69d83d1e47f34ce924c028d6. [Reporting paper]() Recorded evaluation configuration: {"Step":"-","Generator":"Llama-3 8B","Retriever":"bge-m3"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.