We consider automatically identifying the defined term within a mathematical definition from the text of an academic article. Inspired by the development of transformer-based natural language processing applications, we pose the problem as (a) a token-level classification task using fine-tuned pre-trained transformers; and (b) a question-answering task using a generalist large language model (GPT). We also propose a rule-based approach to build a labeled dataset from the LATEX source of papers. Experimental results show that it is possible to reach high levels of precision and recall using either recent (and expensive) GPT 4 or simpler pre-trained models fine-tuned on our task.

In the context of question-answering systems, there are several strategies for scoring candidate answers to definition queries including centroid vectors, bi-term and context language models. These techniques use only positive examples (i.e., descriptions) when building their models. In this work, a maximum entropy based extension is proposed for context language models so as to account for regularities across non-descriptions mined from web-snippets. Experiments show that this extension outperforms other strategies increasing the precision of the top five ranked answers by more than 5%. Results suggest that web-snippets are a cost-efficient source of non-descriptions, and that some relationships extracted from dependency trees are effective to mine for candidate answer sentences.