TikTok has skyrocketed in popularity over recent years, especially among younger audiences. However, there are public concerns about the potential of this platform to promote and amplify harmful content. This study presents a computational analysis of conspiracy theories on TikTok. By leveraging the official TikTok Research API we collect a longitudinal dataset of 1.5M videos shared by creators in the U.S. over three years. We estimate a lower bound on the prevalence of conspiratorial videos (up to 1000 new videos per month) and evaluate platform changes in association with TikTok’s Creativity Program for monetization, observing an overall increase in video duration regardless of content. In addition, we evaluate the capabilities of recent open-weight Large Language Models, both textual and multimodal, to identify conspiracy theories in short vertical format videos. While these models achieve high precision in detecting harmful content (up to $$98\%$$), their overall performance remains comparable to a fine-tuned traditional model such as RoBERTa. Our findings indicate that Large Language Models have the potential to support content moderation strategies on TikTok; however, their deployment involves trade-offs relative to fine-tuned traditional NLP approaches, particularly in terms of performance consistency.
Comparing fine tuned embeddings and large language models for detecting multimodal online conspiracy theories. // Iranian Journal of Science and Technology, Transactions of Electrical Engineering. – 2026. – DOI: 10.1007/s44163-026-02723-w. – URL: https://link.springer.com/article/10.1007/s44163-026-02723-w
(2026). Comparing fine tuned embeddings and large language models for detecting multimodal online conspiracy theories. Iranian Journal of Science and Technology, Transactions of Electrical Engineering. https://doi.org/10.1007/s44163-026-02723-w
“Comparing fine tuned embeddings and large language models for detecting multimodal online conspiracy theories.” Iranian Journal of Science and Technology, Transactions of Electrical Engineering, 2026. doi:10.1007/s44163-026-02723-w.
@article{article35499,
title = {Comparing fine tuned embeddings and large language models for detecting multimodal online conspiracy theories},
journal = {Iranian Journal of Science and Technology, Transactions of Electrical Engineering},
year = {2026},
doi = {10.1007/s44163-026-02723-w},
url = {https://link.springer.com/article/10.1007/s44163-026-02723-w}
}
Background The fashion industry is rapidly transforming with digital technologies and AI. Traditional design evaluation is subjective, inconsistent, and time-consuming. Key challenges include systematically assessing diverse clothing...
Enterprises increasingly combine large language models, retrieval-augmented generation, and autonomous software agents with existing enterprise architecture, knowledge management, and analytics investments, yet these capabilities are...
Generative AI disclosure is commonly treated as a means of making manuscript production transparent. This article argues that its epistemic value is more limited. Authors may accurately report observable operations, retained outputs,...
Automatic detection of voice pathologies from acoustic signals provides a non-invasive and cost-effective solution for early clinical screening. However, developing feature-based detection systems remains challenging under limited...
This paper constructs a comprehensive framework for dance therapy movement segmentation and emotion recognition. This framework primarily addresses the limitations of traditional dance movement emotion assessment. We first use...