# Test-time Reinforcement Learning for Anomalous Video Understanding

arXiv: https://arxiv.org/abs/2609.26099 · HTML: https://mlsanity.com/paper/2609.26099/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 12 of its 24 references that are in the map.
- First posted on arXiv 18 Aug 2026.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)

## Builds on (most cited first, 12 of 12)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Video-LLaVA: Learning United Visual Representation by Alignment Before Projection](https://mlsanity.com/paper/2311.10122/) (2023)
- [LLaVA-Video: Video Instruction Tuning With Synthetic Data](https://mlsanity.com/paper/2410.02713/) (2024)
- [Efficient Test-Time Model Adaptation without Forgetting](https://mlsanity.com/paper/2204.02610/) (2022)
- [InternVideo2: Scaling Foundation Models for Multimodal Video Understanding](https://mlsanity.com/paper/2403.15377/) (2024)
- [Weakly-supervised Video Anomaly Detection with Robust Temporal Feature Magnitude Learning](https://mlsanity.com/paper/2101.10030/) (2021)
- [Holmes-VAU: Towards Long-term Video Anomaly Understanding at Any Granularity](https://mlsanity.com/paper/2412.06171/) (2024)
- [Enhancing Video-LLM Reasoning via Agent-of-Thoughts Distillation](https://mlsanity.com/paper/2412.01694/) (2024)
- [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)
- [Beyond Majority Voting: Towards Fine-grained and More Reliable Reward Signal for Test-Time Reinforcement Learning](https://mlsanity.com/paper/2512.15146/) (2025)
- [Advantage Collapse in Group Relative Policy Optimization: Diagnosis and Mitigation](https://mlsanity.com/paper/2605.21125/) (2026)

## Abstract (arXiv)

Anomalous video understanding aims to identify abnormal events in videos and interpret their semantic meanings beyond simple anomaly detection. Recent video large language models (Video-LLMs) have demonstrated promising zero-shot capabilities for this task, yet their performance remains limited due to insufficient adaptation to diverse anomaly patterns and evolving environments. Test-time reinforcement learning offers a promising solution by enabling models to improve through self-generated feedback signals without requiring additional human annotations. However, applying it to anomalous video understanding remains challenging due to three issues: (1) generated pseudo-labels can be unreliable when consensus is weak; (2) binary reward designs fail to capture uncertainty in model generations, resulting in ineffective optimization signals; and (3) unanimous rollout groups receive identical rewards, causing group-relative advantages to collapse and eliminating effective policy-gradient signals. To address these challenges, we present a novel test-time reinforcement learning framework for anomalous video understanding by introducing dual-query consistency filtering, an entropy-aware consensus reward, and a virtual negative anchor mechanism. The framework retains reliable samples through consistency across semantically equivalent queries, combines answer agreement with generation uncertainty for reward estimation, and introduces a virtual negative anchor to create reward variation in unanimous rollout groups, thereby preserving effective group-relative optimization signals. Experiments on VAU-Bench show that our method outperforms the compared frozen and supervised baselines. The gains are most pronounced on the ECVA subset of VAU-Bench with thinking, where accuracy improves from 75.81% to 90.00% relative to the frozen backbone.