FakeSV-VLM reaches 90.22% and 89.30% accuracy on FakeSV and FakeTT by adding a two-stage MoE adapter and contrastive alignment to InternVL2.5-8B.
The Challenges of Studying Misinformation on Video-Sharing Platforms During Crises and Mass-Convergence Events
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Mis- and disinformation can spread rapidly on video-sharing platforms (VSPs). Despite the growing use of VSPs, there has not been a proportional increase in our ability to understand this medium and the messages conveyed through it. In this work, we draw on our prior experiences to outline three core challenges faced in studying VSPs in high-stakes and fast-paced settings: (1) navigating the unique affordances of VSPs, (2) understanding VSP content and determining its authenticity, and (3) novel user behaviors on VSPs for spreading misinformation. By highlighting these challenges, we hope that researchers can reflect on how to adapt existing research methods and tools to these new contexts, or develop entirely new ones.
fields
cs.MM 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
FakeSV-VLM: Taming VLM for Detecting Fake Short-Video News via Progressive Mixture-Of-Experts Adapter
FakeSV-VLM reaches 90.22% and 89.30% accuracy on FakeSV and FakeTT by adding a two-stage MoE adapter and contrastive alignment to InternVL2.5-8B.