# Text Style Transfer on Yelp (test) Source: [Text Style Transfer on Yelp (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:04:50.810Z Evaluation of models performing Text Style Transfer on the test split of the Yelp dataset, utilizing a comprehensive set of style, fluency, and content preservation metrics. 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 | Text Style Transfer | | Dataset (including any stated split) | Yelp (test) | | Metric | Style Accuracy | | 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 - [Style Accuracy](): Higher is better; unit: PERCENT. - [BLEU (Reference)](): Higher is better; unit: PERCENT. - [BLEU (Self)](): Higher is better; unit: PERCENT. - [Perplexity (PPL)](): Lower is better; unit: SCALAR. - [G-Score](): Higher is better; unit: PERCENT. - [BERT Precision](): Higher is better; unit: PERCENT. - [BERT Recall](): Higher is better; unit: PERCENT. - [BERT F1](): Higher is better; unit: PERCENT. - [Fluency](): Higher is better; unit: PERCENT. - [Content Similarity](): Higher is better; unit: PERCENT. - [Aggregated Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Style Accuracy | Result date | Report paper | | --- | --- | --- | --- | | ARAE | 93.1 | Not recorded | 68fa6c7a1e209131cea342f7 | | RACOLN | 91.3 | Not recorded | 68fa6a7c1e209131cea28842 | | ARAEseq2seq + CONTRA | 89.7 | Not recorded | 68fa6c7a1e209131cea342f7 | | ARAEseq2seq + CONTRA | 89.6 | Not recorded | 68fa6c7a1e209131cea342f7 | | ARAEseq2seq + CLF + CONTRA | 89.4 | Not recorded | 68fa6c7a1e209131cea342f7 | | ARAEseq2seq + CLF + CONTRA | 89.3 | Not recorded | 68fa6c7a1e209131cea342f7 | | ARAEseq2seq | 88.3 | Not recorded | 68fa6c7a1e209131cea342f7 | | Style Transformer | 87.3 | Not recorded | 68fa6a7c1e209131cea28842 | | ARAEseq2seq | 86.7 | Not recorded | 68fa6c7a1e209131cea342f7 | | ARAEseq2seq + CLF | 85.7 | Not recorded | 68fa6c7a1e209131cea342f7 | | ARAEseq2seq + CLF | 85.6 | Not recorded | 68fa6c7a1e209131cea342f7 | | Deep Latent | 85.2 | Not recorded | 68fa6a7c1e209131cea28842 | | ControlledGen | 83.7 | Not recorded | 68fa6a7c1e209131cea28842 | | Cross-Alignment | 74.2 | Not recorded | 68fa6a7c1e209131cea28842 | | DRG | 67.4 | Not recorded | 68fa6c7a1e209131cea342f7 | | ARAE | 2.8 | Not recorded | 68fa6c7a1e209131cea342f7 | | ARAEseq2seq + CLF + CONTRA | 2.8 | Not recorded | 68fa6c7a1e209131cea342f7 | | DRG | 2.3 | Not recorded | 68fa6c7a1e209131cea342f7 | ### ARAE Style Accuracy: 93.1. Result ID: 698f18513a6c314849f5c1ed. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampling":"greedy"}. This record does not assert independent reproduction. ### RACOLN Style Accuracy: 91.3. Result ID: 698e3a7affc3901de09bd452. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ARAEseq2seq + CONTRA Style Accuracy: 89.7. Result ID: 698f18513a6c314849f5c1f3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling":"nucleus(p = 0.6)"}. This record does not assert independent reproduction. ### ARAEseq2seq + CONTRA Style Accuracy: 89.6. Result ID: 698f18513a6c314849f5c1f2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling":"greedy"}. This record does not assert independent reproduction. ### ARAEseq2seq + CLF + CONTRA Style Accuracy: 89.4. Result ID: 698f18513a6c314849f5c1f5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling":"nucleus(p = 0.6)"}. This record does not assert independent reproduction. ### ARAEseq2seq + CLF + CONTRA Style Accuracy: 89.3. Result ID: 698f18513a6c314849f5c1f4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling":"greedy"}. This record does not assert independent reproduction. ### ARAEseq2seq Style Accuracy: 88.3. Result ID: 698f18513a6c314849f5c1ee. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling":"greedy"}. This record does not assert independent reproduction. ### Style Transformer Style Accuracy: 87.3. Result ID: 698e3a7affc3901de09bd450. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ARAEseq2seq Style Accuracy: 86.7. Result ID: 698f18513a6c314849f5c1ef. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling":"nucleus(p = 0.6)"}. This record does not assert independent reproduction. ### ARAEseq2seq + CLF Style Accuracy: 85.7. Result ID: 698f18513a6c314849f5c1f0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling":"greedy"}. This record does not assert independent reproduction. ### ARAEseq2seq + CLF Style Accuracy: 85.6. Result ID: 698f18513a6c314849f5c1f1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling":"nucleus(p = 0.6)"}. This record does not assert independent reproduction. ### Deep Latent Style Accuracy: 85.2. Result ID: 698e3a7affc3901de09bd451. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ControlledGen Style Accuracy: 83.7. Result ID: 698e3a7affc3901de09bd44f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Cross-Alignment Style Accuracy: 74.2. Result ID: 698e3a7affc3901de09bd44e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DRG Style Accuracy: 67.4. Result ID: 698f18513a6c314849f5c1ec. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling":"greedy"}. This record does not assert independent reproduction. ### ARAE Style Accuracy: 2.8. Result ID: 698f18513a6c314849f5c20b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ARAEseq2seq + CLF + CONTRA Style Accuracy: 2.8. Result ID: 698f18513a6c314849f5c20c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DRG Style Accuracy: 2.3. Result ID: 698f18513a6c314849f5c20a. [Reporting paper]() [Method origin paper]() [Implementation]() 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.