Summary of Improving Pseudo Labels with Global-local Denoising Framework For Cross-lingual Named Entity Recognition, by Zhuojun Ding et al.
Improving Pseudo Labels with Global-Local Denoising Framework for Cross-lingual Named Entity Recognition
by Zhuojun Ding, Wei Wei, Xiaoye Qu, Dangyang Chen
First submitted to arxiv on: 3 Jun 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a novel approach for cross-lingual named entity recognition (NER) called Global-Local Denoising framework (GLoDe). The goal is to train an NER model for the target language using only labeled source language data and unlabeled target language data. Current methods introduce noisy labels, leading to performance drops. GLoDe addresses this issue by introducing a progressive denoising strategy that refines pseudo-labeled target language data. Additionally, the paper highlights the importance of target language-specific features, which are often overlooked in previous approaches. Experimental results on two benchmark datasets with six target languages demonstrate that GLoDe outperforms current state-of-the-art methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research aims to help computers recognize named entities (like people and places) in different languages. Currently, methods for this task use labeled data from one language and try to apply it to another language. However, these methods often make mistakes that hurt the results. The new approach, called GLoDe, tries to fix these mistakes by using both general and specific information about the target language. This helps improve the recognition accuracy. The paper shows that this new method works better than existing methods on two sets of data with six different languages. |
Keywords
» Artificial intelligence » Named entity recognition » Ner
