Summary of Toner: Type-oriented Named Entity Recognition with Generative Language Model, by Guochao Jiang et al.
ToNER: Type-oriented Named Entity Recognition with Generative Language Model
by Guochao Jiang, Ziqin Luo, Yuchen Shi, Dixuan Wang, Jiaqing Liang, Deqing Yang
First submitted to arxiv on: 14 Apr 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 A novel named entity recognition (NER) framework, called ToNER, is proposed, leveraging generative models and entity types to improve NER performance. Building upon previous tagging-based or span-based models, ToNER incorporates a type matching model to identify likely entity types in a sentence, followed by fine-tuning the generative model’s encoder through multiple binary classification tasks and an auxiliary task for discovering entity types. Experimental results on NER benchmarks demonstrate the effectiveness of this approach in exploiting entity types’ benefits. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary To better recognize named entities in text, researchers have developed new methods that work well. One way to make it even better is by using information about what kind of entity it is (like person or location). The problem is that we can’t know which ones are important beforehand. To solve this, a team proposes a new approach called ToNER, which uses a special model that can generate text and learn from the data. It does this by identifying the most likely types of entities in a sentence, then refining its understanding to make better predictions. |
Keywords
» Artificial intelligence » Classification » Encoder » Fine tuning » Generative model » Named entity recognition » Ner
