Joint Entity and Relation Extraction on NYT
93.5PrecisionZheng et al.
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Zheng et al.PLM=BERT-base2022.03 | 93.5 | 92.7 | — | 91.9 | — | — | |
| OriginalQuality=-2026.05 | 92.83 | — | — | — | 92.17 | 85.47 | |
| Sui et al.PLM=BERT-base2022.03 | 92.5 | 92.3 | — | 92.2 | — | — | |
| SSDAUQuality=Clean2026.05 | 92 | — | — | — | 92.05 | 85.27 | |
| SSI + SELPLM=T5-v1.1-base2022.03 | 91.94 | 92.6 | — | 93.28 | — | — | |
| SSDAUQuality=Clean2026.05 | 91.74 | — | — | — | 92.9 | 86.74 | |
| Wang et al.PLM=BERT-base2022.03 | 91.4 | 92 | — | 92.6 | — | — | |
| MixupQuality=Clean2026.05 | 91.36 | — | — | — | 90.16 | 82.08 | |
| MixupQuality=Clean2026.05 | 90.56 | — | — | — | 90.06 | 81.92 | |
| OriginalQuality=-2026.05 | 90.17 | — | — | — | 91.45 | 84.24 | |
| WSQuality=Clean2026.05 | 89.91 | — | — | — | 89.61 | 81.17 | |
| NIQuality=Clean2026.05 | 89.37 | — | — | — | 89.91 | 81.67 | |
| BTQuality=Clean2026.05 | 89.1 | — | — | — | 89.54 | 81.07 | |
| SSNQuality=Clean2026.05 | 89.03 | — | — | — | 89.55 | 81.08 | |
| BTQuality=Clean2026.05 | 88.97 | — | — | — | 89.52 | 81.02 | |
| WSQuality=Clean2026.05 | 88.82 | — | — | — | 88.98 | 80.16 | |
| GMQuality=Clean2026.05 | 88.6 | — | — | — | 89.35 | 80.75 | |
| NIQuality=Clean2026.05 | 88.38 | — | — | — | 89.7 | 81.32 | |
| GMQuality=Clean2026.05 | 88.3 | — | — | — | 89.38 | 80.79 | |
| SSNQuality=Clean2026.05 | 88.25 | — | — | — | 89.77 | 81.44 | |
| SSDAUQuality=With Error2026.05 | 83.2 | — | — | — | 80.7 | 67.7 | |
| SSDAUQuality=With Error2026.05 | 80.1 | — | — | — | 77.9 | 63.9 | |
| WSQuality=With Error2026.05 | 78.2 | — | — | — | 71.4 | 55.6 | |
| NIQuality=With Error2026.05 | 77.9 | — | — | — | 69.3 | 53 | |
| MixupQuality=With Error2026.05 | 77.6 | — | — | — | 70.6 | 54.6 | |
| MixupQuality=With Error2026.05 | 76.8 | — | — | — | 75.2 | 60.2 | |
| SSNQuality=With Error2026.05 | 76.1 | — | — | — | 67.2 | 50.6 | |
| WSQuality=With Error2026.05 | 75.6 | — | — | — | 76.3 | 61.8 | |
| GMQuality=With Error2026.05 | 75.3 | — | — | — | 66.2 | 49.5 | |
| NIQuality=With Error2026.05 | 74.8 | — | — | — | 75.2 | 60.2 | |
| SSNQuality=With Error2026.05 | 73.9 | — | — | — | 74.3 | 59.1 | |
| GMQuality=With Error2026.05 | 72.1 | — | — | — | 73.5 | 58.2 | |
| BTQuality=With Error2026.05 | 71.1 | — | — | — | 56.3 | 39.2 | |
| BTQuality=With Error2026.05 | 64.4 | — | — | — | 69.8 | 53.6 | |
| ChatIEQuality=With Error2026.05 | 64.4 | — | — | — | 69.8 | 53.6 | |
| ChatIEQuality=Clean2026.05 | 62.92 | — | — | — | 50.2 | 33.5 | |
| ChatIEQuality=Clean2026.05 | 62.4 | — | — | — | 62.6 | 45.6 | |
| ChatIEQuality=With Error2026.05 | 59.3 | — | — | — | 42.3 | 26.8 | |
| CROSSAGENTIEBackbone=GPT-3.5-Turbo-0125, Prompting Strategy=multi-agent debate, Evaluation Protocol=zero-shot2026.06 | — | 17.23 | 16.77 | — | — | — | |
| DEEPSTRUCTEvaluation protocol=zero-shot2022.05 | — | 60.5 | 28.6 | — | — | — | |
| DEEPSTRUCTEvaluation protocol=multi-task2022.05 | — | 95.4 | 93.7 | — | — | — | |
| DEEPSTRUCTEvaluation protocol=w/ finetune2022.05 | — | 95.9 | 93.3 | — | — | — | |
| RSANTraining Setting=standard2021.01 | — | — | 84.6 | — | — | — | |
| RSAN2022.05 | — | — | 84.6 | — | — | — | |
| SMADE-IEBackbone=GPT-3.5-Turbo-0125, Prompting Strategy=multi-agent debate, Evaluation Protocol=zero-shot2026.06 | — | 30.22 | 29.22 | — | — | — | |
| TANLTraining Setting=standard2021.01 | — | 94.9 | 90.8 | — | — | — | |
| TANLTraining Setting=multi-dataset2021.01 | — | 94.7 | 90.5 | — | — | — | |
| TANLTraining Setting=multi-task2021.01 | — | 94.7 | 90.7 | — | — | — | |
| TANL2022.05 | — | 94.9 | 90.8 | — | — | — | |
| TANL (multitask)mode=multitask2022.05 | — | 94.7 | 90.7 | — | — | — |