# Natural Language Generation on SGD (test) Source: [Natural Language Generation on SGD (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:36:48.617Z Evaluation of Natural Language Generation performance on the test split of the SGD dataset, assessed using BLEU and SER. 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 | Natural Language Generation | | Dataset (including any stated split) | SGD (test) | | Metric | BLEU | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 - [BLEU](): Higher is better; unit: PERCENT. - [SER](): Lower is better; unit: PERCENT. ## Recorded results | Method | BLEU | Result date | Report paper | | --- | --- | --- | --- | | T2G2 | 28.6 | Not recorded | 68fa68ba1e209131cea1dcbc | | T2G2 | 27.8 | Not recorded | 68fa68ba1e209131cea1dcbc | | Naive | 26.3 | Not recorded | 68fa68ba1e209131cea1dcbc | | Schema | 26.2 | Not recorded | 68fa68ba1e209131cea1dcbc | | T2G2 | 26 | Not recorded | 68fa68ba1e209131cea1dcbc | | T2G2 | 24.7 | Not recorded | 68fa68ba1e209131cea1dcbc | | T2G2 | 24.4 | Not recorded | 68fa68ba1e209131cea1dcbc | | T2G2 | 23.8 | Not recorded | 68fa68ba1e209131cea1dcbc | | Naive | 23 | Not recorded | 68fa68ba1e209131cea1dcbc | | Schema | 21.7 | Not recorded | 68fa68ba1e209131cea1dcbc | | Naive | 21.4 | Not recorded | 68fa68ba1e209131cea1dcbc | | Schema | 21.4 | Not recorded | 68fa68ba1e209131cea1dcbc | | Naive | 20.6 | Not recorded | 68fa68ba1e209131cea1dcbc | | Schema | 20.4 | Not recorded | 68fa68ba1e209131cea1dcbc | | Naive | 19.7 | Not recorded | 68fa68ba1e209131cea1dcbc | | Schema | 19.5 | Not recorded | 68fa68ba1e209131cea1dcbc | | Schema | 18.9 | Not recorded | 68fa68ba1e209131cea1dcbc | | Naive | 18.7 | Not recorded | 68fa68ba1e209131cea1dcbc | ### T2G2 BLEU: 28.6. Result ID: 698d80258b517eb43321fa28. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"K":"All"}. This record does not assert independent reproduction. ### T2G2 BLEU: 27.8. Result ID: 698d80258b517eb43321fa27. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"K":"80"}. This record does not assert independent reproduction. ### Naive BLEU: 26.3. Result ID: 698d80258b517eb43321fa1c. [Reporting paper]() Recorded evaluation configuration: {"K":"All"}. This record does not assert independent reproduction. ### Schema BLEU: 26.2. Result ID: 698d80258b517eb43321fa22. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"K":"All"}. This record does not assert independent reproduction. ### T2G2 BLEU: 26. Result ID: 698d80258b517eb43321fa26. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"K":"40"}. This record does not assert independent reproduction. ### T2G2 BLEU: 24.7. Result ID: 698d80258b517eb43321fa25. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"K":"20"}. This record does not assert independent reproduction. ### T2G2 BLEU: 24.4. Result ID: 698d80258b517eb43321fa24. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"K":"10"}. This record does not assert independent reproduction. ### T2G2 BLEU: 23.8. Result ID: 698d80258b517eb43321fa23. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"K":"5"}. This record does not assert independent reproduction. ### Naive BLEU: 23. Result ID: 698d80258b517eb43321fa1b. [Reporting paper]() Recorded evaluation configuration: {"K":"80"}. This record does not assert independent reproduction. ### Schema BLEU: 21.7. Result ID: 698d80258b517eb43321fa21. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"K":"80"}. This record does not assert independent reproduction. ### Naive BLEU: 21.4. Result ID: 698d80258b517eb43321fa1a. [Reporting paper]() Recorded evaluation configuration: {"K":"40"}. This record does not assert independent reproduction. ### Schema BLEU: 21.4. Result ID: 698d80258b517eb43321fa20. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"K":"40"}. This record does not assert independent reproduction. ### Naive BLEU: 20.6. Result ID: 698d80258b517eb43321fa19. [Reporting paper]() Recorded evaluation configuration: {"K":"20"}. This record does not assert independent reproduction. ### Schema BLEU: 20.4. Result ID: 698d80258b517eb43321fa1f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"K":"20"}. This record does not assert independent reproduction. ### Naive BLEU: 19.7. Result ID: 698d80258b517eb43321fa18. [Reporting paper]() Recorded evaluation configuration: {"K":"10"}. This record does not assert independent reproduction. ### Schema BLEU: 19.5. Result ID: 698d80258b517eb43321fa1e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"K":"10"}. This record does not assert independent reproduction. ### Schema BLEU: 18.9. Result ID: 698d80258b517eb43321fa1d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"K":"5"}. This record does not assert independent reproduction. ### Naive BLEU: 18.7. Result ID: 698d80258b517eb43321fa17. [Reporting paper]() Recorded evaluation configuration: {"K":"5"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.