A training-free multi-agent explore-verify system with four specialized agents and a shared evidence registry outperforms zero-shot and RL-finetuned baselines on video anomaly understanding benchmarks.
IEEE Transactions on Image Processing , volume=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
AgenticVAU: Multi-Agent Explore-Verify Reasoning for Video Anomaly Understanding
A training-free multi-agent explore-verify system with four specialized agents and a shared evidence registry outperforms zero-shot and RL-finetuned baselines on video anomaly understanding benchmarks.