# Part-of-Speech Tagging on TWEEBANK V2 (test) Source: [Part-of-Speech Tagging on TWEEBANK V2 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:28:17.826Z Evaluation of Part-of-Speech Tagging performance on the test split of the TWEEBANK V2 dataset, measured using Accuracy. 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 | Part-of-Speech Tagging | | Dataset (including any stated split) | TWEEBANK V2 (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 38 of 38 | | 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 - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | ACE | 95.8 | Not recorded | 68fa6a0e1e209131cea25d83 | | XLM-R-large | 95.5 | Not recorded | 68fa686f1e209131cea1c345 | | XLM-R | 95.4 | Not recorded | 68fa6a0e1e209131cea25d83 | | HuggingFace-BERTweet | 95.38 | Not recorded | 68fa6c821e209131cea3468f | | HuggingFace-BERTweet | 95.21 | Not recorded | 68fa6c821e209131cea3468f | | BERTweet | 95.2 | Not recorded | 68fa686f1e209131cea1c345 | | Nguyen et al. | 95.2 | Not recorded | 68fa6a0e1e209131cea25d83 | | BERTweet | 95.2 | Not recorded | 68fa6c821e209131cea3468f | | XLM-R-large | 95.1 | Not recorded | 68fa686f1e209131cea1c345 | | RoBERTa-large | 94.9 | Not recorded | 68fa686f1e209131cea1c345 | | XLM-R-base | 94.7 | Not recorded | 68fa686f1e209131cea1c345 | | BERTweet | 94.7 | Not recorded | 68fa686f1e209131cea1c345 | | Owoputi et al., 2013 (CRF) | 94.6 | Not recorded | 68fa64411e209131cea03863 | | RoBERTa-large | 94.6 | Not recorded | 68fa686f1e209131cea1c345 | | ARKtagger | 94.6 | Not recorded | 68fa686f1e209131cea1c345 | | Owoputi et al. | 94.6 | Not recorded | 68fa6a0e1e209131cea25d83 | | Owoputi et al. (2013) | 94.6 | Not recorded | 68fa6c821e209131cea3468f | | XLM-R-base | 94.3 | Not recorded | 68fa686f1e209131cea1c345 | | spaCy-XLM-RoBERTa | 93.9 | Not recorded | 68fa6c821e209131cea3468f | | spaCy-XLM-RoBERTa | 93.75 | Not recorded | 68fa6c821e209131cea3468f | | Owoputi et al., 2013 (greedy) | 93.7 | Not recorded | 68fa64411e209131cea03863 | | RoBERTa-base | 93.7 | Not recorded | 68fa686f1e209131cea1c345 | | Owoputi et al. (2013) | 93.7 | Not recorded | 68fa6c821e209131cea3468f | | Stanza | 93.53 | Not recorded | 68fa6c821e209131cea3468f | | RoBERTa-base | 93.5 | Not recorded | 68fa686f1e209131cea1c345 | | Stanza | 93.2 | Not recorded | 68fa6c821e209131cea3468f | | Gui et al. | 92.8 | Not recorded | 68fa6a0e1e209131cea25d83 | | Ma and Hovy, 2016 | 92.5 | Not recorded | 68fa64411e209131cea03863 | | BiLSTM-CNN-CRF | 92.5 | Not recorded | 68fa686f1e209131cea1c345 | | Ma and Hovy (2016) | 92.5 | Not recorded | 68fa6c821e209131cea3468f | | Stanford CoreNLP | 90.6 | Not recorded | 68fa64411e209131cea03863 | | Stanford CoreNLP | 90.6 | Not recorded | 68fa6c821e209131cea3468f | | spaCy | 88.84 | Not recorded | 68fa6c821e209131cea3468f | | FLAIR | 88.19 | Not recorded | 68fa6c821e209131cea3468f | | FLAIR | 87.85 | Not recorded | 68fa6c821e209131cea3468f | | spaCy-BERTweet | 87.61 | Not recorded | 68fa6c821e209131cea3468f | | spaCy | 86.72 | Not recorded | 68fa6c821e209131cea3468f | | spaCy-BERTweet | 86.31 | Not recorded | 68fa6c821e209131cea3468f | ### ACE Accuracy: 95.8. Result ID: 698e0822ffc3901de09a45cd. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Fine-tuning":true,"Trained on both train and development set":true}. This record does not assert independent reproduction. ### XLM-R-large Accuracy: 95.5. Result ID: 698d70188b517eb433218191. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Normalization strategy":"soft"}. This record does not assert independent reproduction. ### XLM-R Accuracy: 95.4. Result ID: 698e0822ffc3901de09a45cc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Fine-tuning":true,"Trained on both train and development set":true}. This record does not assert independent reproduction. ### HuggingFace-BERTweet Accuracy: 95.38. Result ID: 698f1e9e3a6c314849f5f829. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"TB2 + UD English EWT"}. This record does not assert independent reproduction. ### HuggingFace-BERTweet Accuracy: 95.21. Result ID: 698f1e9e3a6c314849f5f828. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"TB2"}. This record does not assert independent reproduction. ### BERTweet Accuracy: 95.2. Result ID: 698d70188b517eb433218197. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Normalization strategy":"soft"}. This record does not assert independent reproduction. ### Nguyen et al. Accuracy: 95.2. Result ID: 698e0822ffc3901de09a45cb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BERTweet Accuracy: 95.2. Result ID: 698f1e9e3a6c314849f5f823. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### XLM-R-large Accuracy: 95.1. Result ID: 698d70188b517eb433218192. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Normalization strategy":"hard"}. This record does not assert independent reproduction. ### RoBERTa-large Accuracy: 94.9. Result ID: 698d70188b517eb43321818f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Normalization strategy":"soft"}. This record does not assert independent reproduction. ### XLM-R-base Accuracy: 94.7. Result ID: 698d70188b517eb433218195. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Normalization strategy":"soft"}. This record does not assert independent reproduction. ### BERTweet Accuracy: 94.7. Result ID: 698d70188b517eb433218198. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Normalization strategy":"hard"}. This record does not assert independent reproduction. ### Owoputi et al., 2013 (CRF) Accuracy: 94.6. Result ID: 698c1aedc330d5bfdbca2b03. [Reporting paper]() Recorded evaluation configuration: {"Tokenization":"gold-standard","Inference strategy":"CRF","Training Data":"UD\_English-EWT + TWEEBANK V2"}. This record does not assert independent reproduction. ### RoBERTa-large Accuracy: 94.6. Result ID: 698d70188b517eb433218190. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Normalization strategy":"hard"}. This record does not assert independent reproduction. ### ARKtagger Accuracy: 94.6. Result ID: 698d70188b517eb433218199. [Reporting paper]() Recorded evaluation configuration: {"Normalization strategy":"hard","Additional training data":"UD\_English-EWT (+c)"}. This record does not assert independent reproduction. ### Owoputi et al. Accuracy: 94.6. Result ID: 698e0822ffc3901de09a45c9. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Owoputi et al. (2013) Accuracy: 94.6. Result ID: 698f1e9e3a6c314849f5f821. [Reporting paper]() Recorded evaluation configuration: {"Decoding":"CRF"}. This record does not assert independent reproduction. ### XLM-R-base Accuracy: 94.3. Result ID: 698d70188b517eb433218196. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Normalization strategy":"hard"}. This record does not assert independent reproduction. ### spaCy-XLM-RoBERTa Accuracy: 93.9. Result ID: 698f1e9e3a6c314849f5f82c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"TB2"}. This record does not assert independent reproduction. ### spaCy-XLM-RoBERTa Accuracy: 93.75. Result ID: 698f1e9e3a6c314849f5f82d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"TB2 + UD English EWT"}. This record does not assert independent reproduction. ### Owoputi et al., 2013 (greedy) Accuracy: 93.7. Result ID: 698c1aedc330d5bfdbca2b02. [Reporting paper]() Recorded evaluation configuration: {"Tokenization":"gold-standard","Inference strategy":"greedy","Training Data":"UD\_English-EWT + TWEEBANK V2"}. This record does not assert independent reproduction. ### RoBERTa-base Accuracy: 93.7. Result ID: 698d70188b517eb433218193. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Normalization strategy":"soft"}. This record does not assert independent reproduction. ### Owoputi et al. (2013) Accuracy: 93.7. Result ID: 698f1e9e3a6c314849f5f820. [Reporting paper]() Recorded evaluation configuration: {"Decoding":"greedy"}. This record does not assert independent reproduction. ### Stanza Accuracy: 93.53. Result ID: 698f1e9e3a6c314849f5f82f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"TB2 + UD English EWT"}. This record does not assert independent reproduction. ### RoBERTa-base Accuracy: 93.5. Result ID: 698d70188b517eb433218194. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Normalization strategy":"hard"}. This record does not assert independent reproduction. ### Stanza Accuracy: 93.2. Result ID: 698f1e9e3a6c314849f5f82e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"TB2"}. This record does not assert independent reproduction. ### Gui et al. Accuracy: 92.8. Result ID: 698e0822ffc3901de09a45ca. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Ma and Hovy, 2016 Accuracy: 92.5. Result ID: 698c1aedc330d5bfdbca2b04. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Tokenization":"gold-standard","Model":"Neural network","Training Data":"UD\_English-EWT + TWEEBANK V2"}. This record does not assert independent reproduction. ### BiLSTM-CNN-CRF Accuracy: 92.5. Result ID: 698d70188b517eb43321819a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Normalization strategy":"hard","Additional training data":"UD\_English-EWT (+c)"}. This record does not assert independent reproduction. ### Ma and Hovy (2016) Accuracy: 92.5. Result ID: 698f1e9e3a6c314849f5f822. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Stanford CoreNLP Accuracy: 90.6. Result ID: 698c1aedc330d5bfdbca2b01. [Reporting paper]() Recorded evaluation configuration: {"Tokenization":"gold-standard","Training Data":"UD\_English-EWT + TWEEBANK V2"}. This record does not assert independent reproduction. ### Stanford CoreNLP Accuracy: 90.6. Result ID: 698f1e9e3a6c314849f5f81f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### spaCy Accuracy: 88.84. Result ID: 698f1e9e3a6c314849f5f825. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"TB2 + UD English EWT"}. This record does not assert independent reproduction. ### FLAIR Accuracy: 88.19. Result ID: 698f1e9e3a6c314849f5f827. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"TB2 + UD English EWT"}. This record does not assert independent reproduction. ### FLAIR Accuracy: 87.85. Result ID: 698f1e9e3a6c314849f5f826. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"TB2"}. This record does not assert independent reproduction. ### spaCy-BERTweet Accuracy: 87.61. Result ID: 698f1e9e3a6c314849f5f82a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"TB2"}. This record does not assert independent reproduction. ### spaCy Accuracy: 86.72. Result ID: 698f1e9e3a6c314849f5f824. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"TB2"}. This record does not assert independent reproduction. ### spaCy-BERTweet Accuracy: 86.31. Result ID: 698f1e9e3a6c314849f5f82b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Data":"TB2 + UD English EWT"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.