# Segment-level Agreement on WMT 2020 (test) Source: [Segment-level Agreement on WMT 2020 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:34:46.667Z Evaluation of model outputs against human judgments regarding segment-level agreement on the WMT 2020 test set, covering multiple English source language pairs. 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 | Segment-level agreement with human ratings | | Dataset (including any stated split) | WMT 2020 (test) | | Metric | Agreement (en-cs) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 7 of 7 | | 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 - [Agreement (en-cs)](): Higher is better; unit: PERCENT. - [Agreement (en-\*)](): Higher is better; unit: PERCENT. - [Agreement (en-de)](): Higher is better; unit: PERCENT. - [Agreement (en-ja)](): Higher is better; unit: PERCENT. - [Agreement (en-pl)](): Higher is better; unit: PERCENT. - [Agreement (en-ru)](): Higher is better; unit: PERCENT. - [Agreement (en-ta)](): Higher is better; unit: PERCENT. - [Agreement (en-zh)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Agreement (en-cs) | Result date | Report paper | | --- | --- | --- | --- | | NANO | 73 | Not recorded | 68fa6ddb1e209131cea3c762 | | NANO | 72.4 | Not recorded | 68fa6ddb1e209131cea3c762 | | BLEURT | 70.8 | Not recorded | 68fa6ddb1e209131cea3c762 | | NANO | 69.8 | Not recorded | 68fa6ddb1e209131cea3c762 | | NANO | 68.2 | Not recorded | 68fa6ddb1e209131cea3c762 | | Prism | 63.8 | Not recorded | 68fa6ddb1e209131cea3c762 | | YiSi | 50.1 | Not recorded | 68fa6ddb1e209131cea3c762 | ### NANO Agreement (en-cs): 73. Result ID: 698fec7a3a6c314849fcd0f9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"XXL","Pretraining stage":"Without (-)"}. This record does not assert independent reproduction. ### NANO Agreement (en-cs): 72.4. Result ID: 698fec7a3a6c314849fcd0fa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"XXL","Pretraining stage":"With (+)"}. This record does not assert independent reproduction. ### BLEURT Agreement (en-cs): 70.8. Result ID: 698fec7a3a6c314849fcd0f4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### NANO Agreement (en-cs): 69.8. Result ID: 698fec7a3a6c314849fcd0f8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"XL","Pretraining stage":"With (+)"}. This record does not assert independent reproduction. ### NANO Agreement (en-cs): 68.2. Result ID: 698fec7a3a6c314849fcd0f7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"XL","Pretraining stage":"Without (-)"}. This record does not assert independent reproduction. ### Prism Agreement (en-cs): 63.8. Result ID: 698fec7a3a6c314849fcd0f5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### YiSi Agreement (en-cs): 50.1. Result ID: 698fec7a3a6c314849fcd0f6. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.