Tiny but Trusted: Efficient Vision-Language Reasoning for Time-Series Anomaly Detection

• URL: http://arxiv.org/abs/2605.30344v1
• Date: Thu, 28 May 2026 17:59:50 GMT
• Title: Tiny but Trusted: Efficient Vision-Language Reasoning for Time-Series Anomaly Detection
• Abstract summary: We develop VisAnomReasoner, a parameter-efficient VLM for time-series anomaly detection.<n>We show that VisAnomReasoner achieves more accurate anomaly localization and consistently outperforms all baselines.<n>Additional experiments on the TSB-AD-U benchmark demonstrate strong cross-benchmark generalization.
• Score: 5.9220868724924856
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent advances in Vision-Language Models (VLMs) have achieved impressive performance across many tasks, yet prior studies report unsatisfactory performance when applying large language or multimodal models to finding abnormal patterns in sequential data. Public anomaly detection benchmarks typically provide interval annotations but not natural-language rationales, making it difficult to fine-tune VLMs to produce grounded, interpretable decisions. To address this gap, we construct VisAnomBench, a curated benchmark built from public time-series datasets and augmented with high-quality anomaly explanations selected from multiple large VLMs using fine-grained, task-specific rewards. Through fine-tuning on this benchmark, we develop VisAnomReasoner, a parameter-efficient VLM for time-series anomaly detection. Experimental results on VisAnomBench show that VisAnomReasoner achieves more accurate anomaly localization and consistently outperforms all baselines, with improvements of at least 21.23 and 23.87 percentage points in precision and F1, respectively. Additional experiments on the TSB-AD-U benchmark demonstrate strong cross-benchmark generalization, with VisAnomReasoner improving precision and F1 by 9.57 and 13.39 percentage points, respectively.

Related papers

• UAVReason: A Unified, Large-Scale Benchmark for Multimodal Aerial Scene Reasoning and Generation [22.090256033688124]
  We introduce UAVReason, the first unified large-scale multi-modal benchmark dedicated to nadir-view UAV scenarios.<n>It consolidates over 273K Visual Question Answering (VQA) pairs, including 23.6K single frames with detailed captions, 68.2K 2-frame temporal sequences, and 188.8K cross-modal generation samples.<n>The benchmark probes 22 diverse reasoning types across spatial and temporal axes while simultaneously evaluating high-fidelity generation across RGB, depth, and segmentation modalities.
  arXiv  Detail & Related papers  (2026-04-07T03:23:30Z)
• SlowFastVAD: Video Anomaly Detection via Integrating Simple Detector and RAG-Enhanced Vision-Language Model [52.47816604709358]
  Video anomaly detection (VAD) aims to identify unexpected events in videos and has wide applications in safety-critical domains.<n> vision-language models (VLMs) have demonstrated strong multimodal reasoning capabilities, offering new opportunities for anomaly detection.<n>We propose SlowFastVAD, a hybrid framework that integrates a fast anomaly detector with a slow anomaly detector.
  arXiv  Detail & Related papers  (2025-04-14T15:30:03Z)
• Inferring Events from Time Series using Language Models [13.414101942484582]
  Time series data measure how environments change over time and drive decision-making in critical domains like finance and healthcare.<n>We conduct the first study of whether Large Language Models (LLMs) can infer events described with natural language from time series data.<n>Several current LLMs demonstrate promising abilities, with OpenAI's o1 performing the best but with DS-R1-distill-Qwen-32B outperforming proprietary models such as GPT-4o.
  arXiv  Detail & Related papers  (2025-03-18T12:07:33Z)
• Debiased Prompt Tuning in Vision-Language Model without Annotations [14.811475313694041]
  Vision-Language Models (VLMs) may suffer from the problem of spurious correlations.<n>By leveraging pseudo-spurious attribute annotations, we propose a method to automatically adjust the training weights of different groups.<n>Our approach efficiently improves the worst-group accuracy on CelebA, Waterbirds, and MetaShift datasets.
  arXiv  Detail & Related papers  (2025-03-11T12:24:54Z)
• GEOBench-VLM: Benchmarking Vision-Language Models for Geospatial Tasks [84.86699025256705]
  We present GEOBench-VLM, a benchmark specifically designed to evaluate Vision-Language Models (VLMs) on geospatial tasks.<n>Our benchmark features over 10,000 manually verified instructions and spanning diverse visual conditions, object types, and scales.<n>We evaluate several state-of-the-art VLMs to assess performance on geospatial-specific challenges.
  arXiv  Detail & Related papers  (2024-11-28T18:59:56Z)
• Uncertainty Aware Learning for Language Model Alignment [97.36361196793929]
  We propose uncertainty-aware learning (UAL) to improve the model alignment of different task scenarios. We implement UAL in a simple fashion -- adaptively setting the label smoothing value of training according to the uncertainty of individual samples. Experiments on widely used benchmarks demonstrate that our UAL significantly and consistently outperforms standard supervised fine-tuning.
  arXiv  Detail & Related papers  (2024-06-07T11:37:45Z)
• Self-supervised Feature Adaptation for 3D Industrial Anomaly Detection [59.41026558455904]
  We focus on multi-modal anomaly detection. Specifically, we investigate early multi-modal approaches that attempted to utilize models pre-trained on large-scale visual datasets. We propose a Local-to-global Self-supervised Feature Adaptation (LSFA) method to finetune the adaptors and learn task-oriented representation toward anomaly detection.
  arXiv  Detail & Related papers  (2024-01-06T07:30:41Z)
• Preserving Knowledge Invariance: Rethinking Robustness Evaluation of Open Information Extraction [49.15931834209624]
  We present the first benchmark that simulates the evaluation of open information extraction models in the real world.<n>We design and annotate a large-scale testbed in which each example is a knowledge-invariant clique.<n>By further elaborating the robustness metric, a model is judged to be robust if its performance is consistently accurate on the overall cliques.
  arXiv  Detail & Related papers  (2023-05-23T12:05:09Z)
• A Closer Look at Debiased Temporal Sentence Grounding in Videos: Dataset, Metric, and Approach [53.727460222955266]
  Temporal Sentence Grounding in Videos (TSGV) aims to ground a natural language sentence in an untrimmed video. Recent studies have found that current benchmark datasets may have obvious moment annotation biases. We introduce a new evaluation metric "dR@n,IoU@m" that discounts the basic recall scores to alleviate the inflating evaluation caused by biased datasets.
  arXiv  Detail & Related papers  (2022-03-10T08:58:18Z)
• NVAE-GAN Based Approach for Unsupervised Time Series Anomaly Detection [19.726089445453734]
  Time series anomaly detection is a common but challenging task in many industries. It is difficult to detect anomalies in time series with high accuracy, due to noisy data collected from real world. We propose our anomaly detection model: Time series to Image VAE (T2IVAE)
  arXiv  Detail & Related papers  (2021-01-08T08:35:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.