Context-Aware Abbreviation Expansion Using Large Language Models

Abstract

Motivated by the need for accelerating text entry in augmentative and alternative communication (AAC) for people with severe motor impairments, we propose a paradigm in which phrases are abbreviated aggressively as primarily word-initial letters. Our approach is to expand the abbreviations into full-phrase options by leveraging conversation context with the power of pretrained large language models (LLMs). Through zero-shot, few-shot, and fine-tuning experiments on four public conversation datasets, we show that for replies to the initial turn of a dialog, an LLM with 64B parameters is able to accurately expand over 70% of phrases with abbreviation length up to 10, leading to an effective keystroke saving rate of up to 77% on these expansions. Including a small amount of context in the form of a single conversation turn more than doubles abbreviation expansion accuracies compared to having no context, an effect that is more pronounced for longer phrases. Additionally, the robustness of the models against typo noise can be enhanced through fine-tuning on noisy data. * equal contribution Related WorkAbbreviation expansion for text entry. Previous research on aiding text entry through AE used abbreviation schemes such as using only content words (Demasco and McCoy, 1992), discarding certain vowels and consonants (Shieber and Nelken, 2007), and flexible letter saving schemes (Pini et al., 2010;Adhikary et al., 2021;Gorman et al., 2021). Spontaneous abbreviations schemes primarily omit vowels, repeating consonants, last characters, and spaces, and lead to modest KSR (e.g., 25-40% in Willis et al. 2005, and 21% in Adhikary et al. 2021.) The low KSR of such schemes can be attributed to the implicit need for a human reader to decode the phrases without significant cognitive burden. N-gram models and neural language models (LMs) have been applied to expanding abbreviations for these relatively low-KSR schemes. By using LSTM models and context, Gorman et al.

NameGuess: Column Name Expansion for Tabular Data

“…Meanwhile, NAMEGUESS is a natural language generation problem suitable for a wide range of lengths of abbreviations and expansions. Regarding abbreviation expansion, our most relevant work is by Cai et al (2022). However, this work primarily addresses text messages/SMS abbreviations, aiming to reduce message length and minimize typos.…”