From Text to Source: Results in Detecting Large Language Model-Generated Content - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2023

From Text to Source: Results in Detecting Large Language Model-Generated Content

Abstract

The widespread use of Large Language Models (LLMs), celebrated for their ability to generate human-like text, has raised concerns about misinformation and ethical implications. Addressing these concerns necessitates the development of robust methods to detect and attribute text generated by LLMs. This paper investigates "Cross-Model Detection," by evaluating whether a classifier trained to distinguish between source LLM-generated and human-written text can also detect text from a target LLM without further training. The study comprehensively explores various LLM sizes and families, and assesses the impact of conversational fine-tuning techniques, quantization, and watermarking on classifier generalization. The research also explores Model Attribution, encompassing source model identification, model family, and model size classification, in addition to quantization and watermarking detection. Our results reveal several key findings: a clear inverse relationship between classifier effectiveness and model size, with larger LLMs being more challenging to detect, especially when the classifier is trained on data from smaller models. Training on data from similarly sized LLMs can improve detection performance from larger models but may lead to decreased performance when dealing with smaller models. Additionally, model attribution experiments show promising results in identifying source models and model families, highlighting detectable signatures in LLM-generated text, with particularly remarkable outcomes in watermarking detection, while no detectable signatures of quantization were observed. Overall, our study contributes valuable insights into the interplay of model size, family, and training data in LLM detection and attribution.
Fichier principal
Vignette du fichier
jqjkvxpwgxhvpqdrgmkzhnygcyyxnqfx.pdf (615.4 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-04264050 , version 1 (27-03-2024)

Licence

Attribution

Identifiers

Cite

Wissam Antoun, Djamé Seddah, Benoît Sagot. From Text to Source: Results in Detecting Large Language Model-Generated Content. The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), May 2024, Torino, Italy. ⟨hal-04264050⟩
69 View
14 Download

Altmetric

Share

Gmail Facebook X LinkedIn More