# Question Answering on HotpotQA (test) (EM and F1 Metrics) Source: [Question Answering on HotpotQA (test) (EM and F1 Metrics)]() Collection index: [llms.txt]() Data updated: 2026-05-01T08:11:53.379Z Evaluation of model performance on the test split of the HotpotQA dataset for Question Answering, reporting Exact Match (EM) and F1 scores. 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) | HotpotQA (test) | | Metric | EM | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 32 of 32 | | 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 - [EM](): Higher is better; unit: PERCENT. - [F1](): Higher is better; unit: PERCENT. ## Recorded results | Method | EM | Result date | Report paper | | --- | --- | --- | --- | | VEROIC | 56.2 | Not recorded | 69f433bb55a3c32af89fd2c0 | | RISK PREDICTOR | 55.9 | Not recorded | 69f433bb55a3c32af89fd2c0 | | VEROIC | 55.8 | Not recorded | 69f433bb55a3c32af89fd2c0 | | CONFIDENCE-THRESHOLD | 55.6 | Not recorded | 69f433bb55a3c32af89fd2c0 | | OUTPUT-TRIGGER | 55.3 | Not recorded | 69f433bb55a3c32af89fd2c0 | | RISK PREDICTOR | 55.3 | Not recorded | 69f433bb55a3c32af89fd2c0 | | CONFIDENCE-THRESHOLD | 54.9 | Not recorded | 69f433bb55a3c32af89fd2c0 | | PERIODIC | 54.7 | Not recorded | 69f433bb55a3c32af89fd2c0 | | OUTPUT-TRIGGER | 54.6 | Not recorded | 69f433bb55a3c32af89fd2c0 | | RANDOM | 53.8 | Not recorded | 69f433bb55a3c32af89fd2c0 | | PERIODIC | 53.3 | Not recorded | 69f433bb55a3c32af89fd2c0 | | RANDOM | 53.1 | Not recorded | 69f433bb55a3c32af89fd2c0 | | FIXED | 51.6 | Not recorded | 69f433bb55a3c32af89fd2c0 | | FIXED | 51.2 | Not recorded | 69f433bb55a3c32af89fd2c0 | | R3AG | 29.13 | Not recorded | 69f03f0c55a3c32af89f9fdb | | IRCoT | 29.05 | Not recorded | 69f03f0c55a3c32af89f9fdb | | Adaptive-RAG | 28.87 | Not recorded | 69f03f0c55a3c32af89f9fdb | | BGE-large | 27.54 | Not recorded | 69f03f0c55a3c32af89f9fdb | | Oracle Single Best | 27.54 | Not recorded | 69f03f0c55a3c32af89f9fdb | | E5-large | 26.87 | Not recorded | 69f03f0c55a3c32af89f9fdb | | Qwen3-embedding-4b | 26.64 | Not recorded | 69f03f0c55a3c32af89f9fdb | | E5-base-v2 | 25.31 | Not recorded | 69f03f0c55a3c32af89f9fdb | | BGE-m3 | 24.73 | Not recorded | 69f03f0c55a3c32af89f9fdb | | Qwen3-embedding-0.6b | 24.42 | Not recorded | 69f03f0c55a3c32af89f9fdb | | RouterRetriever | 24.37 | Not recorded | 69f03f0c55a3c32af89f9fdb | | AAR-contriever-kilt | 23.04 | Not recorded | 69f03f0c55a3c32af89f9fdb | | DIVER-Retriever-0.6b | 22.75 | Not recorded | 69f03f0c55a3c32af89f9fdb | | Random | 22.13 | Not recorded | 69f03f0c55a3c32af89f9fdb | | FLARE | 20.77 | Not recorded | 69f03f0c55a3c32af89f9fdb | | LTRR | 19.86 | Not recorded | 69f03f0c55a3c32af89f9fdb | | Naive | 19.16 | Not recorded | 69f03f0c55a3c32af89f9fdb | | BM25 | 17.87 | Not recorded | 69f03f0c55a3c32af89f9fdb | ### VEROIC EM: 56.2. Result ID: 69f4603e47f34ce924da3302. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Inference Paths":"LLaMA-3.1-8B -\> LLaMA-3.1-70B"}. This record does not assert independent reproduction. ### RISK PREDICTOR EM: 55.9. Result ID: 69f4603e47f34ce924da3301. [Reporting paper]() Recorded evaluation configuration: {"Inference Paths":"LLaMA-3.1-8B -\> LLaMA-3.1-70B"}. This record does not assert independent reproduction. ### VEROIC EM: 55.8. Result ID: 69f4603e47f34ce924da3309. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Inference Paths":"Qwen-2.5-7B -\> Qwen-2.5-32B"}. This record does not assert independent reproduction. ### CONFIDENCE-THRESHOLD EM: 55.6. Result ID: 69f4603e47f34ce924da3300. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Inference Paths":"LLaMA-3.1-8B -\> LLaMA-3.1-70B"}. This record does not assert independent reproduction. ### OUTPUT-TRIGGER EM: 55.3. Result ID: 69f4603e47f34ce924da32ff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Inference Paths":"LLaMA-3.1-8B -\> LLaMA-3.1-70B"}. This record does not assert independent reproduction. ### RISK PREDICTOR EM: 55.3. Result ID: 69f4603e47f34ce924da3308. [Reporting paper]() Recorded evaluation configuration: {"Inference Paths":"Qwen-2.5-7B -\> Qwen-2.5-32B"}. This record does not assert independent reproduction. ### CONFIDENCE-THRESHOLD EM: 54.9. Result ID: 69f4603e47f34ce924da3307. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Inference Paths":"Qwen-2.5-7B -\> Qwen-2.5-32B"}. This record does not assert independent reproduction. ### PERIODIC EM: 54.7. Result ID: 69f4603e47f34ce924da32fe. [Reporting paper]() Recorded evaluation configuration: {"Inference Paths":"LLaMA-3.1-8B -\> LLaMA-3.1-70B"}. This record does not assert independent reproduction. ### OUTPUT-TRIGGER EM: 54.6. Result ID: 69f4603e47f34ce924da3306. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Inference Paths":"Qwen-2.5-7B -\> Qwen-2.5-32B"}. This record does not assert independent reproduction. ### RANDOM EM: 53.8. Result ID: 69f4603e47f34ce924da32fd. [Reporting paper]() Recorded evaluation configuration: {"Inference Paths":"LLaMA-3.1-8B -\> LLaMA-3.1-70B"}. This record does not assert independent reproduction. ### PERIODIC EM: 53.3. Result ID: 69f4603e47f34ce924da3305. [Reporting paper]() Recorded evaluation configuration: {"Inference Paths":"Qwen-2.5-7B -\> Qwen-2.5-32B"}. This record does not assert independent reproduction. ### RANDOM EM: 53.1. Result ID: 69f4603e47f34ce924da3304. [Reporting paper]() Recorded evaluation configuration: {"Inference Paths":"Qwen-2.5-7B -\> Qwen-2.5-32B"}. This record does not assert independent reproduction. ### FIXED EM: 51.6. Result ID: 69f4603e47f34ce924da32fc. [Reporting paper]() Recorded evaluation configuration: {"Inference Paths":"LLaMA-3.1-8B -\> LLaMA-3.1-70B"}. This record does not assert independent reproduction. ### FIXED EM: 51.2. Result ID: 69f4603e47f34ce924da3303. [Reporting paper]() Recorded evaluation configuration: {"Inference Paths":"Qwen-2.5-7B -\> Qwen-2.5-32B"}. This record does not assert independent reproduction. ### R3AG EM: 29.13. Result ID: 69f05bfe47f34ce924d5b5ee. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pipeline Type":"Route"}. This record does not assert independent reproduction. ### IRCoT EM: 29.05. Result ID: 69f05bfe47f34ce924d5b5e9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Pipeline Type":"Loop"}. This record does not assert independent reproduction. ### Adaptive-RAG EM: 28.87. Result ID: 69f05bfe47f34ce924d5b5ea. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pipeline Type":"Conditional"}. This record does not assert independent reproduction. ### BGE-large EM: 27.54. Result ID: 69f05bfe47f34ce924d5b5e0. [Reporting paper]() Recorded evaluation configuration: {"Pipeline Type":"Sequential"}. This record does not assert independent reproduction. ### Oracle Single Best EM: 27.54. Result ID: 69f05bfe47f34ce924d5b5e7. [Reporting paper]() Recorded evaluation configuration: {"Pipeline Type":"Conditional"}. This record does not assert independent reproduction. ### E5-large EM: 26.87. Result ID: 69f05bfe47f34ce924d5b5df. [Reporting paper]() Recorded evaluation configuration: {"Pipeline Type":"Sequential"}. This record does not assert independent reproduction. ### Qwen3-embedding-4b EM: 26.64. Result ID: 69f05bfe47f34ce924d5b5e3. [Reporting paper]() Recorded evaluation configuration: {"Pipeline Type":"Sequential"}. This record does not assert independent reproduction. ### E5-base-v2 EM: 25.31. Result ID: 69f05bfe47f34ce924d5b5de. [Reporting paper]() Recorded evaluation configuration: {"Pipeline Type":"Sequential"}. This record does not assert independent reproduction. ### BGE-m3 EM: 24.73. Result ID: 69f05bfe47f34ce924d5b5e1. [Reporting paper]() Recorded evaluation configuration: {"Pipeline Type":"Sequential"}. This record does not assert independent reproduction. ### Qwen3-embedding-0.6b EM: 24.42. Result ID: 69f05bfe47f34ce924d5b5e2. [Reporting paper]() Recorded evaluation configuration: {"Pipeline Type":"Sequential"}. This record does not assert independent reproduction. ### RouterRetriever EM: 24.37. Result ID: 69f05bfe47f34ce924d5b5ed. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pipeline Type":"Route"}. This record does not assert independent reproduction. ### AAR-contriever-kilt EM: 23.04. Result ID: 69f05bfe47f34ce924d5b5eb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Pipeline Type":"Sequential"}. This record does not assert independent reproduction. ### DIVER-Retriever-0.6b EM: 22.75. Result ID: 69f05bfe47f34ce924d5b5e4. [Reporting paper]() Recorded evaluation configuration: {"Pipeline Type":"Sequential"}. This record does not assert independent reproduction. ### Random EM: 22.13. Result ID: 69f05bfe47f34ce924d5b5e6. [Reporting paper]() Recorded evaluation configuration: {"Pipeline Type":"Conditional"}. This record does not assert independent reproduction. ### FLARE EM: 20.77. Result ID: 69f05bfe47f34ce924d5b5e8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pipeline Type":"Loop"}. This record does not assert independent reproduction. ### LTRR EM: 19.86. Result ID: 69f05bfe47f34ce924d5b5ec. [Reporting paper]() Recorded evaluation configuration: {"Pipeline Type":"Route"}. This record does not assert independent reproduction. ### Naive EM: 19.16. Result ID: 69f05bfe47f34ce924d5b5dd. [Reporting paper]() Recorded evaluation configuration: {"Pipeline Type":"–"}. This record does not assert independent reproduction. ### BM25 EM: 17.87. Result ID: 69f05bfe47f34ce924d5b5e5. [Reporting paper]() Recorded evaluation configuration: {"Pipeline Type":"Sequential"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.