Real-world surveillance often operates in low-light environments, where degraded visual evidence can make training-free anomaly reasoning unreliable. However, current training-free methods typically assume sufficiently clean inputs, which can lead to hallucinated semantics and unstable anomaly scores under visual degradation. To address this issue, we propose Retrieval-augmented Contextual Inference (ReCI), a training-free framework that leverages retrieval-augmented context for robust low-light video anomaly detection. ReCI constructs Semantic Context (SC) through hierarchical captioning by aggregating clip-level local captions into a video-level global description. It then performs contextualized anomaly inference using Reference Context (RC) retrieved from a reference pool built during inference. The VLM outputs anomaly scores and associated confidence values, which we use as a heuristic reliability signal for temporal refinement. Experiments on XD-Violence and UCF-Crime show that ReCI consistently improves over prior training-free baselines, with particularly clear gains on the low-light subset.

Existing weakly supervised video violence detection (VVD) methods primarily rely on Euclidean representation learning, which often struggles to distinguish visually similar yet semantically distinct events due to limited hierarchical modeling and insufficient ambiguous training samples. To address this challenge, we propose PiercingEye, a novel dual-space learning framework that synergizes Euclidean and hyperbolic geometries to enhance discriminative feature representation. Specifically, PiercingEye introduces a layer-sensitive hyperbolic aggregation strategy with hyperbolic Dirichlet energy constraints to progressively model event hierarchies, and a cross-space attention mechanism to facilitate complementary feature interactions between Euclidean and hyperbolic spaces. Furthermore, to mitigate the scarcity of ambiguous samples, we leverage large language models to generate logic-guided ambiguous event descriptions, enabling explicit supervision through a hyperbolic vision-language contrastive loss that prioritizes high-confusion samples via dynamic similarity-aware weighting. Extensive experiments on XD-Violence and UCF-Crime benchmarks demonstrate that PiercingEye achieves state-of-the-art performance, with particularly strong results on a newly curated ambiguous event subset, validating its superior capability in fine-grained violence detection.

While unmanned aerial vehicles (UAVs) offer wide-area, high-altitude coverage for anomaly detection, they face challenges such as dynamic viewpoints, scale variations, and complex scenes. Existing datasets and methods, mainly designed for fixed ground-level views, struggle to adapt to these conditions, leading to significant performance drops in drone-view scenarios.To bridge this gap, we introduce A2Seek (Aerial Anomaly Seek), a large-scale, reasoning-centric benchmark dataset for aerial anomaly understanding. This dataset covers various scenarios and environmental conditions, providing high-resolution real-world aerial videos with detailed annotations, including anomaly categories, frame-level timestamps, region-level bounding boxes, and natural language explanations for causal reasoning. Building on this dataset, we propose A2Seek-R1, a novel reasoning framework that generalizes R1-style strategies to aerial anomaly understanding, enabling a deeper understanding of “Where” anomalies occur and “Why” they happen in aerial frames.To this end, A2Seek-R1 first employs a graph-of-thought (GoT)-guided supervised fine-tuning approach to activate the model's latent reasoning capabilities on A2Seek. Then, we introduce Aerial Group Relative Policy Optimization (A-GRPO) to design rule-based reward functions tailored to aerial scenarios. Furthermore, we propose a novel “seeking” mechanism that simulates UAV flight behavior by directing the model's attention to informative regions.Extensive experiments demonstrate that A2Seek-R1 achieves up to a 22.04\% improvement in AP for prediction accuracy and a 13.9\% gain in mIoU for anomaly localization, exhibiting strong generalization across complex environments and out-of-distribution scenarios. Our dataset and code are released at https://2-mo.github.io/A2Seek/.

Recent blind super-resolution (BSR) methods are explored to handle unknown degradations and achieve impressive performance. However, the prevailing assumption in most BSR methods is the spatial invariance of degradation kernels across the entire image, which leads to significant performance declines when faced with spatially variant degradations caused by object motion or defocusing. Additionally, these methods do not account for the human visual system's tendency to focus differently on areas of varying perceptual difficulty, as they uniformly process each pixel during reconstruction. To cope with these issues, we propose a difficulty-guided variant degradation learning network for BSR, named difficulty-guided degradation learning (DDL)-BSR, which explores the relationship between reconstruction difficulty and degradation estimation. Accordingly, the proposed DDL-BSR consists of three customized networks: reconstruction difficulty prediction (RDP), space-variant degradation estimation (SDE), and degradation and difficulty-informed reconstruction (DDR). Specifically, RDP learns the reconstruction difficulty with the proposed reconstruction-distance supervision. Then, SDE is designed to estimate space-variant degradation kernels according to the difficulty map. Finally, both degradation kernels and reconstruction difficulty are fed into DDR, which takes into account such two prior knowledge information to guide super-resolution (SR). Experimental analysis on various synthetic datasets demonstrates that DDL-BSR invariably surpasses state-of-the-art (SOTA) methods, producing SR images with enhanced realism and texture quality. Code is available at https://github.com/JiaWang0704/DDL-BSR.

In this paper, we propose a novel framework, Modality-Free Violence Detection (MFVD), which captures the causal relationships among multimodal cues and ensures stable performance even in the absence of audio information. Specifically, we design a novel Cross-Modal Causal Attention mechanism (CCA) to deal with modality asynchrony by utilizing relative temporal distance and semantic correlation to obtain causal attention between audio and visual information instead of merely calculating correlation scores between audio and visual features. Moreover, to ensure our framework can work well when the audio modality is missing, we design a Cross-Modal Feature Distillation module (CFD), leveraging the common parts of the fused features obtained from CCA to guide the enhancement of visual features. Experimental results on the XD-Violence dataset demonstrate the superior performance of the proposed method in both vision-only and audio-visual modalities, surpassing state-of-the-art methods for both tasks.

In this paper, we propose Dual Space Embedding Learning (DSEL) for weakly supervised audio-visual violence detection, which excavates violence information deeply in both Euclidean and Hyperbolic spaces to distinguish violence from non-violence semantically and alleviate the asynchronous issue of violent cues in audio-visual patterns. Specifically, we first design a dual space visual feature interaction module (DSVFI) to deeply investigate the violence information in visual modality, which contains richer information compared to audio counterpart. Then, considering the modality asynchrony between the two modalities, we employ a late modality fusion method and design an asynchrony-aware audio-visual fusion module (AAF), in which visual features receive the violent prompt from the audio features after interacting among snippets and learning the violence information from each other. Experimental results show that our method achieves state-of-the-art performance on XD-Violence.

Drone-view Object Detection (DOD) is a meaningful but challenging task. It hits a bottleneck due to two main reasons: (1) The high proportion of difficult objects (e.g., small objects, occluded objects, etc.) makes the detection performance unsatisfactory. (2) The unevenly distributed objects make detection inefficient. These two factors also lead to a phenomenon, obeying the Pareto principle, that some challenging regions occupying a low area proportion of the image have a significant impact on the final detection while the vanilla regions occupying the major area have a negligible impact due to the limited room for performance improvement. Motivated by the human visual system that naturally attempts to invest unequal energies in things of hierarchical difficulty for recognizing objects effectively, this paper presents a novel Pareto Refocusing Detection (PRDet) network that distinguishes the challenging regions from the vanilla regions under reverse-attention guidance and refocuses the challenging regions with the assistance of the region-specific context. Specifically, we first propose a Reverse-attention Exploration Module (REM) that excavates the potential position of difficult objects by suppressing the features which are salient to the commonly used detector. Then, we propose a Region-specific Context Learning Module (RCLM) that learns to generate specific contexts for strengthening the understanding of challenging regions. It is noteworthy that the specific context is not shared globally but unique for each challenging region with the exploration of spatial and appearance cues. Extensive experiments and comprehensive evaluations on the VisDrone2021-DET and UAVDT datasets demonstrate that the proposed PRDet can effectively improve the detection performance, especially for those difficult objects, outperforming state-of-the-art detectors. Furthermore, our method also achieves significant performance improvements on the DTU-Drone dataset for power inspection.