# Long-form Question Answering on Fava Source: [Long-form Question Answering on Fava]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:08:13.702Z Evaluation of model performance on the Fava dataset for Long-form Question Answering, measured using VeriScore F1 and AlpacaEval Win Rate. 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 | Long-form Question Answering | | Dataset (including any stated split) | Fava | | Metric | VeriScore F1 | | 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 - [VeriScore F1](): Higher is better; unit: PERCENT. - [AlpacaEval Win Rate](): Higher is better; unit: PERCENT. ## Recorded results | Method | VeriScore F1 | Result date | Report paper | | --- | --- | --- | --- | | EWE | 61 | Not recorded | 68fa79081e209131cea7bc9f | | DRAGIN | 57.2 | Not recorded | 68fa79081e209131cea7bc9f | | RA | 56.8 | Not recorded | 68fa79081e209131cea7bc9f | | EWE | 53.1 | Not recorded | 68fa79081e209131cea7bc9f | | COVE w/ Retrieval | 52.6 | Not recorded | 68fa79081e209131cea7bc9f | | Llama-3.1 | 52 | Not recorded | 68fa79081e209131cea7bc9f | | RA | 51.8 | Not recorded | 68fa79081e209131cea7bc9f | | DRAGIN | 51.1 | Not recorded | 68fa79081e209131cea7bc9f | | Llama-3.1 | 51 | Not recorded | 68fa79081e209131cea7bc9f | | NEST | 50.3 | Not recorded | 68fa79081e209131cea7bc9f | | NEST | 50.2 | Not recorded | 68fa79081e209131cea7bc9f | | COVE | 49.5 | Not recorded | 68fa79081e209131cea7bc9f | | COVE w/ Retrieval | 39.5 | Not recorded | 68fa79081e209131cea7bc9f | | COVE | 38.7 | Not recorded | 68fa79081e209131cea7bc9f | ### EWE VeriScore F1: 61. Result ID: 6995bc933badfaffe6be575a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 70B"}. This record does not assert independent reproduction. ### DRAGIN VeriScore F1: 57.2. Result ID: 6995bc933badfaffe6be5757. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 70B"}. This record does not assert independent reproduction. ### RA VeriScore F1: 56.8. Result ID: 6995bc933badfaffe6be5755. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 70B"}. This record does not assert independent reproduction. ### EWE VeriScore F1: 53.1. Result ID: 6995bc933badfaffe6be5761. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 8B"}. This record does not assert independent reproduction. ### COVE w/ Retrieval VeriScore F1: 52.6. Result ID: 6995bc933badfaffe6be5759. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 70B"}. This record does not assert independent reproduction. ### Llama-3.1 VeriScore F1: 52. Result ID: 6995bc933badfaffe6be5754. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 70B"}. This record does not assert independent reproduction. ### RA VeriScore F1: 51.8. Result ID: 6995bc933badfaffe6be575c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 8B"}. This record does not assert independent reproduction. ### DRAGIN VeriScore F1: 51.1. Result ID: 6995bc933badfaffe6be575e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 8B"}. This record does not assert independent reproduction. ### Llama-3.1 VeriScore F1: 51. Result ID: 6995bc933badfaffe6be575b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 8B"}. This record does not assert independent reproduction. ### NEST VeriScore F1: 50.3. Result ID: 6995bc933badfaffe6be5756. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 70B"}. This record does not assert independent reproduction. ### NEST VeriScore F1: 50.2. Result ID: 6995bc933badfaffe6be575d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 8B"}. This record does not assert independent reproduction. ### COVE VeriScore F1: 49.5. Result ID: 6995bc933badfaffe6be5758. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 70B"}. This record does not assert independent reproduction. ### COVE w/ Retrieval VeriScore F1: 39.5. Result ID: 6995bc933badfaffe6be5760. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 8B"}. This record does not assert independent reproduction. ### COVE VeriScore F1: 38.7. Result ID: 6995bc933badfaffe6be575f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.1 8B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.