Study on relationship between adversarial texts and language errors: a human-computer interaction perspective

Rui Xiao, Kuangyi Zhang, Yunchun Zhang, Qi Wang, Shangyu Hou and Ling Liu

Behaviour and Information Technology, 2025, vol. 44, issue 9, 2053-2068

Abstract: Large language models (LLMs) are widely applied in many human-computer interactive applications, such as chatbots. However, a deep understanding of the vulnerability of LLMs against adversarial attacks and language errors poses a threatening challenge. This study, therefore, presents a systematic analysis to disclose the relationships among language errors, adversarial texts, and LLMs. Each LLM is measured in language understanding ability and robustness within a human-computer interaction context. To further disclose the differences between language errors and adversarial texts, we measured each LLM under 6 metrics, including Levinstein edit distance, Modification rate, Cosine similarity, perplexity, error rate, and BLEU. Through detailed experiments, we first prove that both language error texts and adversarial texts have a serious impact on the performance of LLMs. The quantified measure of the difference between these two texts is innovative in differentiating language errors and adversarial texts from clean texts.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/0144929X.2024.2396437 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:tbitxx:v:44:y:2025:i:9:p:2053-2068

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/tbit20

DOI: 10.1080/0144929X.2024.2396437

Access Statistics for this article

Behaviour and Information Technology is currently edited by Dr Panos P Markopoulos

More articles in Behaviour and Information Technology from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().