Summary of Modeling Citation Worthiness by Using Attention-based Bidirectional Long Short-term Memory Networks and Interpretable Models, By Tong Zeng et al.
Modeling citation worthiness by using attention-based bidirectional long short-term memory networks and interpretable models
by Tong Zeng, Daniel E. Acuna
First submitted to arxiv on: 20 May 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Machine Learning (cs.LG)
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 This paper proposes a machine learning approach to automatically detect sentences that require citations in scientific texts, aiming to improve the robustness and well-construction of scientific arguments. The authors employ a Bidirectional Long Short-Term Memory (BiLSTM) network with attention mechanism and contextual information to classify sentences as citation-worthy or not. A new large dataset, PMOA-CITE, is constructed from open access publications in PubMed’s Open Access Subset, which outperforms previous datasets. The proposed architecture achieves state-of-the-art performance on the standard ACL-ARC dataset (F1=0.507) and high performance (F1=0.856) on the new PMOA-CITE dataset. Additionally, the paper explores interpretable models to understand how specific language is used to promote or inhibit citations, revealing that sections and surrounding sentences are crucial for improved predictions. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Scientists often struggle with citing sources in their research papers. Sometimes they forget or don’t know where to put a citation. This paper tries to solve this problem by using machine learning to automatically detect when a sentence needs a citation. The researchers use a new way of analyzing sentences called BiLSTM and attention mechanism. They also create a large dataset from open access scientific articles. The results show that their approach is very accurate in detecting when a sentence needs a citation. This can help scientists make better arguments and make it easier for others to understand their research. |
Keywords
» Artificial intelligence » Attention » Machine learning
