# Machine Translation Meta-evaluation on WMT EN-ZH 2025 Source: [Machine Translation Meta-evaluation on WMT EN-ZH 2025]() Collection index: [llms.txt]() Data updated: 2026-05-12T11:16:34.840Z Evaluation of machine translation quality using meta-evaluation metrics (such as acc\*eq and SPA) on the WMT English-to-Chinese 2025 test set. 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 | Machine Translation Meta-evaluation | | Dataset (including any stated split) | WMT EN-ZH 2025 | | Metric | Acc\*Eq | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | 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 - [Acc\*Eq](): Higher is better; unit: PERCENT. - [SPA](): Higher is better; unit: PERCENT. ## Recorded results | Method | Acc\*Eq | Result date | Report paper | | --- | --- | --- | --- | | MM (MLP) | 56.8 | Not recorded | 6a02b45bdbc836b87d8249ca | | MM (MLP) | 56.7 | Not recorded | 6a02b45bdbc836b87d8249ca | | DMM (MLP) | 56.7 | Not recorded | 6a02b45bdbc836b87d8249ca | | DMM (MLP) | 56.7 | Not recorded | 6a02b45bdbc836b87d8249ca | | MM (GP) | 55.4 | Not recorded | 6a02b45bdbc836b87d8249ca | | DMM (GP) | 55.3 | Not recorded | 6a02b45bdbc836b87d8249ca | | MM (GP) | 54.8 | Not recorded | 6a02b45bdbc836b87d8249ca | | wmt22-comet-da | 54.4 | Not recorded | 6a02b45bdbc836b87d8249ca | | DMM (CL) | 54.2 | Not recorded | 6a02b45bdbc836b87d8249ca | | DMM (CL) | 53.7 | Not recorded | 6a02b45bdbc836b87d8249ca | | DMM (GP) | 53.1 | Not recorded | 6a02b45bdbc836b87d8249ca | | Gemini-2.5-Pro | 53 | Not recorded | 6a02b45bdbc836b87d8249ca | | MM (OLS) | 51 | Not recorded | 6a02b45bdbc836b87d8249ca | | MM (OLS) | 47.3 | Not recorded | 6a02b45bdbc836b87d8249ca | | DMM (OLS) | 46.3 | Not recorded | 6a02b45bdbc836b87d8249ca | | DMM (OLS) | 44.9 | Not recorded | 6a02b45bdbc836b87d8249ca | | gemba | 35.3 | Not recorded | 6a02b45bdbc836b87d8249ca | ### MM (MLP) Acc\*Eq: 56.8. Result ID: 6a030c0347f34ce924e6c946. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Pooled","Model Variant":"MM","Combiner":"MLP"}. This record does not assert independent reproduction. ### MM (MLP) Acc\*Eq: 56.7. Result ID: 6a030c0347f34ce924e6c949. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Separate","Model Variant":"MM","Combiner":"MLP"}. This record does not assert independent reproduction. ### DMM (MLP) Acc\*Eq: 56.7. Result ID: 6a030c0347f34ce924e6c94c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Pooled","Model Variant":"DMM","Combiner":"MLP","k":"6"}. This record does not assert independent reproduction. ### DMM (MLP) Acc\*Eq: 56.7. Result ID: 6a030c0347f34ce924e6c950. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Separate","Model Variant":"DMM","Combiner":"MLP","k":"6"}. This record does not assert independent reproduction. ### MM (GP) Acc\*Eq: 55.4. Result ID: 6a030c0347f34ce924e6c947. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Separate","Model Variant":"MM","Combiner":"GP"}. This record does not assert independent reproduction. ### DMM (GP) Acc\*Eq: 55.3. Result ID: 6a030c0347f34ce924e6c94a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Pooled","Model Variant":"DMM","Combiner":"GP","k":"6"}. This record does not assert independent reproduction. ### MM (GP) Acc\*Eq: 54.8. Result ID: 6a030c0347f34ce924e6c944. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Pooled","Model Variant":"MM","Combiner":"GP"}. This record does not assert independent reproduction. ### wmt22-comet-da Acc\*Eq: 54.4. Result ID: 6a030c0347f34ce924e6c942. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DMM (CL) Acc\*Eq: 54.2. Result ID: 6a030c0347f34ce924e6c94d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Pooled","Model Variant":"DMM","Combiner":"CL","k":"6"}. This record does not assert independent reproduction. ### DMM (CL) Acc\*Eq: 53.7. Result ID: 6a030c0347f34ce924e6c951. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Separate","Model Variant":"DMM","Combiner":"CL","k":"6"}. This record does not assert independent reproduction. ### DMM (GP) Acc\*Eq: 53.1. Result ID: 6a030c0347f34ce924e6c94e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Separate","Model Variant":"DMM","Combiner":"GP","k":"6"}. This record does not assert independent reproduction. ### Gemini-2.5-Pro Acc\*Eq: 53. Result ID: 6a030c0347f34ce924e6c941. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MM (OLS) Acc\*Eq: 51. Result ID: 6a030c0347f34ce924e6c948. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Separate","Model Variant":"MM","Combiner":"OLS"}. This record does not assert independent reproduction. ### MM (OLS) Acc\*Eq: 47.3. Result ID: 6a030c0347f34ce924e6c945. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Pooled","Model Variant":"MM","Combiner":"OLS"}. This record does not assert independent reproduction. ### DMM (OLS) Acc\*Eq: 46.3. Result ID: 6a030c0347f34ce924e6c94b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Pooled","Model Variant":"DMM","Combiner":"OLS","k":"6"}. This record does not assert independent reproduction. ### DMM (OLS) Acc\*Eq: 44.9. Result ID: 6a030c0347f34ce924e6c94f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pooling Strategy":"Language-Separate","Model Variant":"DMM","Combiner":"OLS","k":"6"}. This record does not assert independent reproduction. ### gemba Acc\*Eq: 35.3. Result ID: 6a030c0347f34ce924e6c943. [Reporting paper]() [Method origin 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.