Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

DiffusionNER: Boundary Diffusion for Named Entity Recognition

About

In this paper, we propose DiffusionNER, which formulates the named entity recognition task as a boundary-denoising diffusion process and thus generates named entities from noisy spans. During training, DiffusionNER gradually adds noises to the golden entity boundaries by a fixed forward diffusion process and learns a reverse diffusion process to recover the entity boundaries. In inference, DiffusionNER first randomly samples some noisy spans from a standard Gaussian distribution and then generates the named entities by denoising them with the learned reverse diffusion process. The proposed boundary-denoising diffusion process allows progressive refinement and dynamic sampling of entities, empowering DiffusionNER with efficient and flexible entity generation capability. Experiments on multiple flat and nested NER datasets demonstrate that DiffusionNER achieves comparable or even better performance than previous state-of-the-art models.

Yongliang Shen, Kaitao Song, Xu Tan, Dongsheng Li, Weiming Lu, Yueting Zhuang• 2023

Related benchmarks

TaskDatasetResultRank
Nested Named Entity RecognitionACE 2004 (test)
F1 Score88.39
166
Nested Named Entity RecognitionACE 2005 (test)
F1 Score86.93
153
Nested Named Entity RecognitionGENIA (test)
F1 Score85.62
140
Named Entity RecognitionCoNLL 03
F1 (Entity)92.78
102
Named Entity RecognitionOntoNotes
F1-score90.66
91
Named Entity RecognitionMSRA
F1 Score94.91
29
Showing 6 of 6 rows

Other info

Code

Follow for update