Single Word Change Is All You Need: Using LLMs to Create Synthetic Training Examples for Text Classifiers

Overview of attention for article published in Expert Systems, July 2025

Altmetric Badge

About this Attention Score

In the top 25% of all research outputs scored by Altmetric
One of the highest-scoring outputs from this source (#9 of 347)
High Attention Score compared to outputs of the same age (92nd percentile)
High Attention Score compared to outputs of the same age and source (85th percentile)

Mentioned by

news: 2 news outlets
twitter: 7 X users

reddit: 1 Redditor
bluesky: 2 Bluesky users

Readers on

mendeley: 2 Mendeley

So far, Altmetric has seen 8 X posts from 7 X users, with an upper bound of 1,043,701 followers.

RT @TechXplore_com: #A new software suite uses large language models to identify and defend against single-word attacks that can fool AI te…

#A new software suite uses large language models to identify and defend against single-word attacks that can fool AI text classifiers, significantly improving their reliability and robustness. @MIT https://t.co/FsYNalHNid https://t.co/oOZZcn5S4R

RT @TechXplore_com: #A new software suite uses large language models to identify and defend against single-word attacks that can fool AI te…

This page shows the most recent X posts that mention this research output. Click here to find out how to access more activity, including 4 additional X posts.