# Learning to Watch: Active Video Anomaly Understanding via Interleaved Policy Optimization

arXiv: https://arxiv.org/abs/2607.00622 · HTML: https://mlsanity.com/paper/2607.00622/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 33 of its 37 references that are in the map.
- First posted on arXiv 1 Jul 2026.

## Builds on (most cited first, 30 of 33)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [Real-world Anomaly Detection in Surveillance Videos](https://mlsanity.com/paper/1801.04264/) (2018)
- [Not only Look, but also Listen: Learning Multimodal Violence Detection under Weak Supervision](https://mlsanity.com/paper/2007.04687/) (2020)
- [Weakly-supervised Video Anomaly Detection with Robust Temporal Feature Magnitude Learning](https://mlsanity.com/paper/2101.10030/) (2021)
- [VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection](https://mlsanity.com/paper/2308.11681/) (2023)
- [UBnormal: New Benchmark for Supervised Open-Set Video Anomaly Detection](https://mlsanity.com/paper/2111.08644/) (2021)
- [Dual Memory Units with Uncertainty Regulation for Weakly Supervised Video Anomaly Detection](https://mlsanity.com/paper/2302.05160/) (2023)
- [Harnessing Large Language Models for Training-free Video Anomaly Detection](https://mlsanity.com/paper/2404.01014/) (2024)
- [Holmes-VAU: Towards Long-term Video Anomaly Understanding at Any Granularity](https://mlsanity.com/paper/2412.06171/) (2024)
- [Follow the Rules: Reasoning for Video Anomaly Detection with Large Language Models](https://mlsanity.com/paper/2407.10299/) (2024)
- [Hawk: Learning to Understand Open-World Video Anomalies](https://mlsanity.com/paper/2405.16886/) (2024)
- [Uncovering What, Why and How: A Comprehensive Benchmark for Causation Understanding of Video Anomaly](https://mlsanity.com/paper/2405.00181/) (2024)
- [Self-Distilled Masked Auto-Encoders are Efficient Video Anomaly Detectors](https://mlsanity.com/paper/2306.12041/) (2023)
- [EventVAD: Training-Free Event-Aware Video Anomaly Detection](https://mlsanity.com/paper/2504.13092/) (2025)
- [Vad-R1: Towards Video Anomaly Reasoning via Perception-to-Cognition Chain-of-Thought](https://mlsanity.com/paper/2505.19877/) (2025)
- [VAU-R1: Advancing Video Anomaly Understanding via Reinforcement Fine-Tuning](https://mlsanity.com/paper/2505.23504/) (2025)
- [PANDA: Towards Generalist Video Anomaly Detection via Agentic AI Engineer](https://mlsanity.com/paper/2509.26386/) (2025)
- [VADTree: Explainable Training-Free Video Anomaly Detection via Hierarchical Granularity-Aware Tree](https://mlsanity.com/paper/2510.22693/) (2025)
- [IAD-R1: Reinforcing Consistent Reasoning in Industrial Anomaly Detection](https://mlsanity.com/paper/2508.09178/) (2025)
- [A Unified Reasoning Framework for Holistic Zero-Shot Video Anomaly Analysis](https://mlsanity.com/paper/2511.00962/) (2025)
- [Shape-centered Representation Learning for Visible-Infrared Person Re-identification](https://mlsanity.com/paper/2310.17952/) (2023)
- [Video-Level Language-Driven Video-Based Visible-Infrared Person Re-Identification](https://mlsanity.com/paper/2506.02439/) (2025)
- [Aligning Effective Tokens with Video Anomaly in Large Language Models](https://mlsanity.com/paper/2508.06350/) (2025)
- [A2Seek: Towards Reasoning-Centric Benchmark for Aerial Anomaly Understanding](https://mlsanity.com/paper/2505.21962/) (2025)

## Built on by (most cited first, 4 of 4)

- [AgenticVAU: Multi-Agent Explore-Verify Reasoning for Video Anomaly Understanding](https://mlsanity.com/paper/2608.03779/) (2026)
- [Representation--Behavior Alignment for Explainable Weakly-Supervised Video Anomaly Detection](https://mlsanity.com/paper/2610.05129/) (2026)
- [RoMod: Temporal Routing Modulation via Mixture-of-Experts for Video Anomaly Detection](https://mlsanity.com/paper/2610.05131/) (2026)
- [Anomaly-Guided Observation and Two-Level Semantic Feedback for Training-Free Video Anomaly Analysis](https://mlsanity.com/paper/2610.16324/) (2026)

## Abstract (arXiv)

Video anomaly understanding (VAU) relies on sparse, context-dependent cues. However, existing passive paradigms suffer from observational aliasing, where static sampling fails to disambiguate semantically distinct events. To overcome this, we propose Anom-π, a closed-loop framework that reconceptualizes video understanding as an active sequential decision-making process within a dynamic environment. Inspired by human video-reviewing behavior, this framework unifies internal cognitive reasoning and strategic evidence acquisition into an interleaved policy, utilizing temporal atomic operators such as local backtracking, temporal expansion, and fine-grained sampling to endow the model with perceptual proactivity. To learn such complex interaction strategies under video-level weak supervision, we design Interactive Direct Preference Optimization (iDPO) to achieve trajectory-level policy alignment, guided by an Active Evidence Inquiry (AEI) utility that balances task success, informative evidence acquisition, and interaction cost. This approach enables the agent to learn to actively disambiguate hypotheses while suppressing redundant exploration. Extensive experiments demonstrate that our framework, with only 2B parameters, achieves highly competitive performance, significantly outperforming state-of-the-art large-scale VAU models in complex scenarios.