Open-Domain, Content-based, Multi-modal Fact-checking of Out-of-Context Images via Online Resources
원문 초록
Misinformation is now a major problem due to its potential high risks to our core democratic and societal values and orders. Out-of-context misinformation is one of the easiest and effective ways used by adversaries to spread viral false stories. In this threat, a real image is re-purposed to support other narratives by misrepresenting its context and/or elements. The internet is being used as the go-to way to verify information using different sources and modalities. Our goal is an inspectable method that automates this time-consuming and reasoning-intensive process by fact-checking the image-caption pairing using Web evidence. To integrate evidence and cues from both modalities, we introduce the concept of 'multi-modal cycle-consistency check'; starting from the image/caption, we gather textual/visual evidence, which will be compared against the other paired caption/image, respectively. Moreover, we propose a novel architecture, Consistency-Checking Network (CCN), that mimics the layered human reasoning across the same and different modalities: the caption vs. textual evidence, the image vs. visual evidence, and the image vs. caption. Our work offers the first step and benchmark for open-domain, content-based, multi-modal fact-checking, and significantly outperforms previous baselines that did not leverage external evidence.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
RealFactBench: A Benchmark for Evaluating Large Language Models in Real-World Fact-Checking
SUCEA: Reasoning-Intensive Retrieval for Adversarial Fact-checking through Claim Decomposition and Editing
Entity-aware Cross-lingual Claim Detection for Automated Fact-checking
Veracity: An Open-Source AI Fact-Checking System
In Crowd Veritas: Leveraging Human Intelligence To Fight Misinformation
Combating Misinformation in the Arab World: Challenges & Opportunities
토론
이 논문과 연결된 포럼 글이 아직 없습니다.