EVID-Bench supplies 222 videos across nine manipulation types in three categories and shows that frontier multimodal models reach at most 61.43% point-level accuracy when forced to use web search to identify false information.
In2021 IEEE in- ternational conference on big data (big data), pages 899–908
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
citing papers explorer
-
When Seeing Is Not Believing -- A Benchmark for Search-Grounded Video Misinformation Detection
EVID-Bench supplies 222 videos across nine manipulation types in three categories and shows that frontier multimodal models reach at most 61.43% point-level accuracy when forced to use web search to identify false information.
- Probing Multimodal Large Language Models on Cognitive Biases in Chinese Short-Video Misinformation