# Data-to-text generation on SPNLG (test) Source: [Data-to-text generation on SPNLG (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:36:47.632Z Evaluation of data-to-text generation performance on the test split of the SPNLG dataset, assessed using metrics including BLEU, NIST, METEOR, ROUGE, CIDEr, and Self-BLEU. 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 | Data-to-text generation | | Dataset (including any stated split) | SPNLG (test) | | Metric | BLEU | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 9 of 9 | | 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. - [NIST](): Higher is better; unit: PERCENT. - [METEOR](): Higher is better; unit: PERCENT. - [ROUGE](): Higher is better; unit: PERCENT. - [CIDEr](): Higher is better; unit: SCALAR. - [Self-BLEU](): Lower is better; unit: PERCENT. ## Recorded results | Method | BLEU | Result date | Report paper | | --- | --- | --- | --- | | Table2seq-beam | 40.61 | Not recorded | 68fa67081e209131cea13fc1 | | Table2seq-pretrain | 40.56 | Not recorded | 68fa67081e209131cea13fc1 | | VTM | 40.04 | Not recorded | 68fa67081e209131cea13fc1 | | VTM-noraw | 39.94 | Not recorded | 68fa67081e209131cea13fc1 | | VTM-noraw | 39.63 | Not recorded | 68fa67081e209131cea13fc1 | | VTM | 39.58 | Not recorded | 68fa67081e209131cea13fc1 | | VTM-noraw | 38.33 | Not recorded | 68fa67081e209131cea13fc1 | | Table2seq-sample | 34.97 | Not recorded | 68fa67081e209131cea13fc1 | | Temp-KN | 6.45 | Not recorded | 68fa67081e209131cea13fc1 | ### Table2seq-beam BLEU: 40.61. Result ID: 698d16048b517eb4331ee5ca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding Strategy":"beam search"}. This record does not assert independent reproduction. ### Table2seq-pretrain BLEU: 40.56. Result ID: 698d16048b517eb4331ee5cc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pre-training":true}. This record does not assert independent reproduction. ### VTM BLEU: 40.04. Result ID: 698d16048b517eb4331ee5ce. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Raw Data Usage":true}. This record does not assert independent reproduction. ### VTM-noraw BLEU: 39.94. Result ID: 698d16048b517eb4331ee5d0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Raw Data Usage":false}. This record does not assert independent reproduction. ### VTM-noraw BLEU: 39.63. Result ID: 698d16048b517eb4331ee5d2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Objective":"w/o LMI - Lpt","Raw Data Usage":false}. This record does not assert independent reproduction. ### VTM BLEU: 39.58. Result ID: 698d16048b517eb4331ee5cf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Objective":"w/o Lpc","Raw Data Usage":true}. This record does not assert independent reproduction. ### VTM-noraw BLEU: 38.33. Result ID: 698d16048b517eb4331ee5d1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Objective":"w/o LMI","Raw Data Usage":false}. This record does not assert independent reproduction. ### Table2seq-sample BLEU: 34.97. Result ID: 698d16048b517eb4331ee5cb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding Strategy":"sampling"}. This record does not assert independent reproduction. ### Temp-KN BLEU: 6.45. Result ID: 698d16048b517eb4331ee5cd. [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.