Researchers introduced AV-FABLE, an audio-visual manipulation forensics benchmark for long-form videos, arguing that video-level classification is insufficient because localized AI manipulations typically affect brief events within otherwise authentic content. The work targets a gap the authors identify: existing audio-visual deepfake benchmarks largely focus on short talking-head clips rather than long-form temporal forensics. The benchmark matters for forensic analysis, where detecting and localizing short manipulated segments in long recordings is the core task.
Localized AI manipulations in long-form videos often affect brief events within otherwise authentic content, making video-level classification insufficient for forensic analysis. Existing audio-visual deepfake benchmarks largely focus on short talking-head clips, while long-form temporal forensics r
EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.