# Relation Extraction on TACRED v1.0 (test) Source: [Relation Extraction on TACRED v1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:43:13.432Z Evaluation of model performance on the test split of the TACRED v1.0 dataset for Relation Extraction, typically measured using F1, Precision, and Recall. 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 | Relation Extraction | | Dataset (including any stated split) | TACRED v1.0 (test) | | Metric | F1 Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 37 of 37 | | 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 - [F1 Score](): Higher is better; unit: PERCENT. - [Precision](): Higher is better; unit: PERCENT. - [Recall](): Higher is better; unit: PERCENT. ## Recorded results | Method | F1 Score | Result date | Report paper | | --- | --- | --- | --- | | RoBERTa\_LARGE + typed entity marker (punct) | 91.1 | Not recorded | 68fa6d881e209131cea3a6b1 | | RoBERTa\_LARGE + entity marker | 90.5 | Not recorded | 68fa6d881e209131cea3a6b1 | | LUKE | 90.3 | Not recorded | 68fa6d881e209131cea3a6b1 | | BERT\_LARGE + typed entity marker | 89.7 | Not recorded | 68fa6d881e209131cea3a6b1 | | BERT\_LARGE + entity marker | 89.2 | Not recorded | 68fa6d881e209131cea3a6b1 | | BERT\_BASE + typed entity marker | 87.9 | Not recorded | 68fa6d881e209131cea3a6b1 | | BERT\_BASE + entity marker | 87.7 | Not recorded | 68fa6d881e209131cea3a6b1 | | SpanBERT | 85.3 | Not recorded | 68fa6d881e209131cea3a6b1 | | RoBERTa\_LARGE + typed entity marker (punct) | 83.2 | Not recorded | 68fa6d881e209131cea3a6b1 | | BERT\_LARGE + typed entity marker | 81.3 | Not recorded | 68fa6d881e209131cea3a6b1 | | RoBERTa\_LARGE + entity marker | 81.2 | Not recorded | 68fa6d881e209131cea3a6b1 | | LUKE | 80.6 | Not recorded | 68fa6d881e209131cea3a6b1 | | C-GCN | 80.3 | Not recorded | 68fa6d881e209131cea3a6b1 | | PA-LSTM | 79.4 | Not recorded | 68fa6d881e209131cea3a6b1 | | KnowBERT | 79.3 | Not recorded | 68fa6d881e209131cea3a6b1 | | BERT\_BASE + typed entity marker | 79.3 | Not recorded | 68fa6d881e209131cea3a6b1 | | SpanBERT | 78 | Not recorded | 68fa6d881e209131cea3a6b1 | | BERT\_LARGE + entity marker | 77.9 | Not recorded | 68fa6d881e209131cea3a6b1 | | BERT\_BASE + entity marker | 77.2 | Not recorded | 68fa6d881e209131cea3a6b1 | | C-GCN | 74.6 | Not recorded | 68fa6d881e209131cea3a6b1 | | PA-LSTM | 73.3 | Not recorded | 68fa6d881e209131cea3a6b1 | | SimpleRE | 71.7 | Not recorded | 68fa6f211e209131cea43e16 | | KnowBERT | 71.5 | Not recorded | 68fa6f211e209131cea43e16 | | SpanBERT | 70.8 | Not recorded | 68fa6f211e209131cea43e16 | | GDPNet | 70.5 | Not recorded | 68fa6f211e209131cea43e16 | | LST-AGCN | 68.8 | Not recorded | 68fa6f211e209131cea43e16 | | C-AGGCN | 68.2 | Not recorded | 68fa6f211e209131cea43e16 | | PLM | 66.9 | Not recorded | 68fa6c121e209131cea31e9d | | PA-LSTM | 65.1 | Not recorded | 68fa6f211e209131cea43e16 | | LSTM | 62.7 | Not recorded | 68fa6f211e209131cea43e16 | | PRBOOST | 48.1 | Not recorded | 68fa6c121e209131cea31e9d | | COSINE | 41 | Not recorded | 68fa6c121e209131cea31e9d | | Snorkel | 40.8 | Not recorded | 68fa6c121e209131cea31e9d | | Snorkel | 39.7 | Not recorded | 68fa6c121e209131cea31e9d | | COSINE | 39.5 | Not recorded | 68fa6c121e209131cea31e9d | | PLM | 32.9 | Not recorded | 68fa6c121e209131cea31e9d | | Rule Matching | 20.1 | Not recorded | 68fa6c121e209131cea31e9d | ### RoBERTa\_LARGE + typed entity marker (punct) F1 Score: 91.1. Result ID: 698fa7f23a6c314849fade92. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"RoBERTa-Large","Entity Representation":"typed entity marker (punct)"}. This record does not assert independent reproduction. ### RoBERTa\_LARGE + entity marker F1 Score: 90.5. Result ID: 698fa7f23a6c314849fade8d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"RoBERTa-Large","Entity Representation":"entity marker"}. This record does not assert independent reproduction. ### LUKE F1 Score: 90.3. Result ID: 698fa7f23a6c314849fade8f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LUKE"}. This record does not assert independent reproduction. ### BERT\_LARGE + typed entity marker F1 Score: 89.7. Result ID: 698fa7f23a6c314849fade91. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"BERT-Large","Entity Representation":"typed entity marker"}. This record does not assert independent reproduction. ### BERT\_LARGE + entity marker F1 Score: 89.2. Result ID: 698fa7f23a6c314849fade8c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"BERT-Large","Entity Representation":"entity marker"}. This record does not assert independent reproduction. ### BERT\_BASE + typed entity marker F1 Score: 87.9. Result ID: 698fa7f23a6c314849fade90. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"BERT-Base","Entity Representation":"typed entity marker"}. This record does not assert independent reproduction. ### BERT\_BASE + entity marker F1 Score: 87.7. Result ID: 698fa7f23a6c314849fade8b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"BERT-Base","Entity Representation":"entity marker"}. This record does not assert independent reproduction. ### SpanBERT F1 Score: 85.3. Result ID: 698fa7f23a6c314849fade8e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"SpanBERT"}. This record does not assert independent reproduction. ### RoBERTa\_LARGE + typed entity marker (punct) F1 Score: 83.2. Result ID: 698fa7f23a6c314849fade88. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"RoBERTa-Large","Entity Representation":"typed entity marker (punct)"}. This record does not assert independent reproduction. ### BERT\_LARGE + typed entity marker F1 Score: 81.3. Result ID: 698fa7f23a6c314849fade87. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"BERT-Large","Entity Representation":"typed entity marker"}. This record does not assert independent reproduction. ### RoBERTa\_LARGE + entity marker F1 Score: 81.2. Result ID: 698fa7f23a6c314849fade82. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"RoBERTa-Large","Entity Representation":"entity marker"}. This record does not assert independent reproduction. ### LUKE F1 Score: 80.6. Result ID: 698fa7f23a6c314849fade85. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LUKE"}. This record does not assert independent reproduction. ### C-GCN F1 Score: 80.3. Result ID: 698fa7f23a6c314849fade8a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Architecture":"Sequence-based"}. This record does not assert independent reproduction. ### PA-LSTM F1 Score: 79.4. Result ID: 698fa7f23a6c314849fade89. [Reporting paper]() Recorded evaluation configuration: {"Architecture":"Sequence-based"}. This record does not assert independent reproduction. ### KnowBERT F1 Score: 79.3. Result ID: 698fa7f23a6c314849fade84. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"KnowBERT"}. This record does not assert independent reproduction. ### BERT\_BASE + typed entity marker F1 Score: 79.3. Result ID: 698fa7f23a6c314849fade86. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"BERT-Base","Entity Representation":"typed entity marker"}. This record does not assert independent reproduction. ### SpanBERT F1 Score: 78. Result ID: 698fa7f23a6c314849fade83. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"SpanBERT"}. This record does not assert independent reproduction. ### BERT\_LARGE + entity marker F1 Score: 77.9. Result ID: 698fa7f23a6c314849fade81. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"BERT-Large","Entity Representation":"entity marker"}. This record does not assert independent reproduction. ### BERT\_BASE + entity marker F1 Score: 77.2. Result ID: 698fa7f23a6c314849fade80. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"BERT-Base","Entity Representation":"entity marker"}. This record does not assert independent reproduction. ### C-GCN F1 Score: 74.6. Result ID: 698fa7f23a6c314849fade7f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Architecture":"Sequence-based"}. This record does not assert independent reproduction. ### PA-LSTM F1 Score: 73.3. Result ID: 698fa7f23a6c314849fade7e. [Reporting paper]() Recorded evaluation configuration: {"Architecture":"Sequence-based"}. This record does not assert independent reproduction. ### SimpleRE F1 Score: 71.7. Result ID: 6990a79b3a6c314849022e19. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"BERT"}. This record does not assert independent reproduction. ### KnowBERT F1 Score: 71.5. Result ID: 6990a79b3a6c314849022e1a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Knowledge Source":"External Knowledge Base"}. This record does not assert independent reproduction. ### SpanBERT F1 Score: 70.8. Result ID: 6990a79b3a6c314849022e17. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GDPNet F1 Score: 70.5. Result ID: 6990a79b3a6c314849022e18. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"SpanBERT"}. This record does not assert independent reproduction. ### LST-AGCN F1 Score: 68.8. Result ID: 6990a79b3a6c314849022e16. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### C-AGGCN F1 Score: 68.2. Result ID: 6990a79b3a6c314849022e15. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### PLM F1 Score: 66.9. Result ID: 698ec6b0ffc3901de0a10295. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"supervision":"Supervised","training\_data":"100%"}. This record does not assert independent reproduction. ### PA-LSTM F1 Score: 65.1. Result ID: 6990a79b3a6c314849022e14. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LSTM F1 Score: 62.7. Result ID: 6990a79b3a6c314849022e13. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### PRBOOST F1 Score: 48.1. Result ID: 698ec6b0ffc3901de0a1029c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"supervision":"Weakly Supervised"}. This record does not assert independent reproduction. ### COSINE F1 Score: 41. Result ID: 698ec6b0ffc3901de0a1029b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"supervision":"Weakly Supervised","fine-tuning":"on clean data"}. This record does not assert independent reproduction. ### Snorkel F1 Score: 40.8. Result ID: 698ec6b0ffc3901de0a1029a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"supervision":"Weakly Supervised","fine-tuning":"on clean data"}. This record does not assert independent reproduction. ### Snorkel F1 Score: 39.7. Result ID: 698ec6b0ffc3901de0a10298. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"supervision":"Weakly Supervised"}. This record does not assert independent reproduction. ### COSINE F1 Score: 39.5. Result ID: 698ec6b0ffc3901de0a10299. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"supervision":"Weakly Supervised"}. This record does not assert independent reproduction. ### PLM F1 Score: 32.9. Result ID: 698ec6b0ffc3901de0a10296. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"supervision":"Supervised","training\_data":"limited"}. This record does not assert independent reproduction. ### Rule Matching F1 Score: 20.1. Result ID: 698ec6b0ffc3901de0a10297. [Reporting paper]() Recorded evaluation configuration: {"supervision":"Weakly Supervised"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.