Artificial Text Detection via Examining the Topology of Attention Maps - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès ACL Anthology Année : 2021

Artificial Text Detection via Examining the Topology of Attention Maps

Résumé

The impressive capabilities of recent generative models to create texts that are challenging to distinguish from the human-written ones can be misused for generating fake news, product reviews, and even abusive content. Despite the prominent performance of existing methods for artificial text detection, they still lack interpretability and robustness towards unseen models. To this end, we propose three novel types of interpretable topological features for this task based on Topological Data Analysis (TDA) which is currently understudied in the field of NLP. We empirically show that the features derived from the BERT model outperform count-and neural-based baselines up to 10% on three common datasets, and tend to be the most robust towards unseen GPT-style generation models as opposed to existing methods. The probing analysis of the features reveals their sensitivity to the surface and syntactic properties. The results demonstrate that TDA is a promising line with respect to NLP tasks, specifically the ones that incorporate surface and structural information.
Fichier principal
Vignette du fichier
2021.emnlp-main.50.pdf (961.56 Ko) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte

Dates et versions

hal-03456191 , version 1 (21-12-2021)

Identifiants

  • HAL Id : hal-03456191 , version 1

Citer

Laida Kushnareva, Daniil Cherniavskii, Vladislav Mikhailov, Ekaterina Artemova, Serguei Barannikov, et al.. Artificial Text Detection via Examining the Topology of Attention Maps. Empirical Methods in Natural Language Processing, ACL (Association for Computational Linguistics), Nov 2021, Punta Cana, Dominican Republic. ⟨hal-03456191⟩
60 Consultations
60 Téléchargements

Partager

Gmail Facebook X LinkedIn More