Research map / Large language models / Retrieval and knowledge
Hermes: An Evidence-Driven Agentic Framework for Trustworthy and Explainable AI-Generated Video Detection
Shuaibo Li, Pengfei HAO, Hongtao Wu, Jianfeng Dong, Ping Li, Xiaohong Liu, Lei Zhu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add RAG-based instance-conditioned strategy planning (ablation) | Hermes · GenVideo | AUROC: 95.12 -> 97.84 | Improved |
| add Evidence Reasoning Graph organization (ablation) | Hermes · GenVideo | AUROC: 95.94 -> 97.84 | Improved |
| add graph-grounded multi-agent deliberation (ablation) | Hermes · GenVideo | AUROC: 96.21 -> 97.84 | Improved |
| add deliberation rounds from 1 to 3 (ablation) | Hermes · GenVideo | AUROC: 96.83 -> 97.84 | Improved |
| increase deliberation rounds from 3 to 4 (ablation) | Hermes · GenVideo | AUROC: 97.84 -> 97.79 | No clear effect |
| increase RAG top-k from 1 to 5 (ablation) | Hermes · GenVideo | AUROC: 96.03 -> 97.84 | Improved |
| increase RAG top-k from 5 to 7 (ablation) | Hermes · GenVideo | AUROC: 97.84 -> 97.21 | Hurt |
| use Hermes agentic framework instead of D3 (training-free baseline) | AI-generated video detection · GenVideo | FNR: 8.93 -> 4.87 | Improved |
| use Hermes without training instead of supervised detectors (NSG-VD) | AI-generated video detection · GenVideo | AUROC: 96.14 -> 97.84 | Improved |
Closest papers by meaning
- SafeGuard: A Multi-Agent Perception-Reasoning Framework for Social-Risk AI-Generated Video Detection (ECCV 2026 · similarity 0.89)
- REVEAL: Reasoning-Enhanced Forensic Evidence Analysis for Explainable AI-Generated Image Detection (ECCV 2026 · similarity 0.89)
- BusterX: MLLM-Powered AI-Generated Video Forgery Detection and Explanation (NeurIPS 2026 · similarity 0.87)
- Seeing What Matters: Generalizable AI-generated Video Detection with Forensic-Oriented Augmentation (NeurIPS 2025 · similarity 0.85)
- LOKI: A Comprehensive Synthetic Data Detection Benchmark using Large Multimodal Models (ICLR 2025 · similarity 0.85)
- Locate-Then-Examine: Grounded Region Reasoning Improves Detection of AI-Generated Images (CVPR 2026 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
