Perception, Understanding and Reasoning: A Multimodal Benchmark for Video Fake News Detection

The advent of multi-modal large language models (MLLMs) has greatly advanced research on video fake news detection (VFND) tasks. Existing benchmarks typically focus on the detection accuracy, while failing to provide finegrained assessments for the entire detection process. To address these limitations, we introduce POVFNDB (Process-oriented Video Fake News Detection Benchmark), a processoriented benchmark comprising 10 tasks designed to systematically evaluate MLLMs' perception, understanding, and reasoning capabilities in VFND. This benchmark contains 36,240 human-annotated question-answer (QA) in structured or open-ended formats, spanning 15 distinct evaluation dimensions that characterize different aspects of the video fake news detection process. Using POVFNDB, we conduct comprehensive evaluations on both proprietary and open-source MLLMs. Moreover, we establish a strong benchmark baseline by fine-tuning Qwen2.5VL-7B-Instruct on process-oriented chain-of-thought data constructed with our proposed POVFND-CoT framework, achieving state-of-the-art performance on VFND. MLLMs Perception …on-screen text 'Russian Airshow …the on-site footage in the video …German Air Force Base Understanding …the video shows an airshow scene.… Focus on entire process ( ENHR, KHR, RHR, ACC) Classification Model CORRECT Focus on final result ( Accuracy, F1) Reasoning …Text-Video mismatch…Fake