HintsOfTruth: A Multimodal Checkworthiness Detection Dataset with Real and Synthetic Claims

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 <PRE_TAG>synthetic data</POST_TAG> 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 <PRE_TAG>multimodal models</POST_TAG> but the first 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 <PRE_TAG>synthetic data</POST_TAG>, <PRE_TAG>multimodal models</POST_TAG> perform more robustly