Researchers have developed a new system called NVKE-CEI to combat fake news in short videos. This system addresses limitations in existing methods by improving keyframe selection and integrating both content analysis and external evidence retrieval. NVKE-CEI utilizes a novel keyframe extraction technique and a framework with two LLM-based fact-checkers, which are then combined by a judge model to provide explanations for fake news videos. Experiments demonstrate that NVKE-CEI surpasses current state-of-the-art approaches in detecting fake news and generating accurate, content-grounded explanations. AI
IMPACT This research could lead to more effective tools for identifying and explaining misinformation on video platforms.
RANK_REASON The item is an academic paper detailing a new method for detecting fake news. [lever_c_demoted from research: ic=1 ai=1.0]
- alphaXiv
- arXiv
- CatalyzeX Code Finder for Papers
- Connected Papers
- CORE Recommender
- DagsHub
- Gotit.pub
- Hugging Face
- Litmaps
- NVKE
- NVKE-CEI
- ScienceCast
- scite Smart Citations
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
