Saved in:
Bibliographic Details
Main Authors: van der Meer, Michiel, Korshunov, Pavel, Marcel, Sébastien, van der Plas, Lonneke
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.11753
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918045138550784
author van der Meer, Michiel
Korshunov, Pavel
Marcel, Sébastien
van der Plas, Lonneke
author_facet van der Meer, Michiel
Korshunov, Pavel
Marcel, Sébastien
van der Plas, Lonneke
contents Misinformation can be countered with fact-checking, but the process is costly and slow. Identifying checkworthy claims is the first step, where automation can help scale fact-checkers' efforts. However, detection methods struggle with content that is (1) multimodal, (2) from diverse domains, and (3) synthetic. We introduce HintsOfTruth, a public dataset for multimodal checkworthiness detection with 27K real-world and synthetic image/claim pairs. The mix of real and synthetic data makes this dataset unique and ideal for benchmarking detection methods. We compare fine-tuned and prompted Large Language Models (LLMs). We find that well-configured lightweight text-based encoders perform comparably to multimodal models but the former only focus on identifying non-claim-like content. Multimodal LLMs can be more accurate but come at a significant computational cost, making them impractical for large-scale applications. When faced with synthetic data, multimodal models perform more robustly.
format Preprint
id arxiv_https___arxiv_org_abs_2502_11753
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle HintsOfTruth: A Multimodal Checkworthiness Detection Dataset with Real and Synthetic Claims
van der Meer, Michiel
Korshunov, Pavel
Marcel, Sébastien
van der Plas, Lonneke
Artificial Intelligence
Misinformation can be countered with fact-checking, but the process is costly and slow. Identifying checkworthy claims is the first step, where automation can help scale fact-checkers' efforts. However, detection methods struggle with content that is (1) multimodal, (2) from diverse domains, and (3) synthetic. We introduce HintsOfTruth, a public dataset for multimodal checkworthiness detection with 27K real-world and synthetic image/claim pairs. The mix of real and synthetic data makes this dataset unique and ideal for benchmarking detection methods. We compare fine-tuned and prompted Large Language Models (LLMs). We find that well-configured lightweight text-based encoders perform comparably to multimodal models but the former only focus on identifying non-claim-like content. Multimodal LLMs can be more accurate but come at a significant computational cost, making them impractical for large-scale applications. When faced with synthetic data, multimodal models perform more robustly.
title HintsOfTruth: A Multimodal Checkworthiness Detection Dataset with Real and Synthetic Claims
topic Artificial Intelligence
url https://arxiv.org/abs/2502.11753