mlsanityData to 7 Oct 2026

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

arXiv 2605.30344 ↗First posted 28 May 2026cs.AIcited by 0

On the map: LLM reasoning → Chain-of-thought reasoning for driving

also in 1 more
Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Chain-of-thought reasoning for driving, one of 17 directions in LLM reasoning. Builds on 8 of its 62 references that are in the map. First posted on arXiv 28 May 2026.

Builds on

8 of 62 references.

References · 8

  1. 2025Qwen2.5-VL Technical Reportcited by 5,418
  2. 2025Gemma 3 Technical Reportcited by 1,686
  3. 2021An Evaluation of Anomaly Detection and Diagnosis in Multivariate Time Seriescited by 68
  4. 2025Time Series Forecasting via Reasoning: A Slow-Thinking Approach with Reinforcement Fine-Tuned LLMscited by 34
  5. 2025TempoGPT: Enhancing Time Series Reasoning via Quantizing Embeddingcited by 7
  6. 2025Time-RA: Towards Time Series Reasoning for Anomaly Diagnosis with LLM Feedbackcited by 7
  7. 2025mTSBench: Benchmarking Multivariate Time Series Anomaly Detection and Model Selection at Scalecited by 6
  8. 2025AXIS: Explainable Time Series Anomaly Detection with Large Language Modelscited by 4

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Harnessing Vision-Language Models for Time Series Anomaly Detection
  2. 2024Quo Vadis, Anomaly Detection? LLMs and VLMs in the Spotlight
  3. 2025ViTs: Teaching Machines to See Time Series Anomalies Like Human Experts
  4. 2026Analyzing VLM-Based Approaches for Anomaly Classification and Segmentation
  5. 2024Vision-Language Models Assisted Unsupervised Video Anomaly Detection
  6. 2025From Benchmarks to Reality: Advancing Visual Anomaly Detection by the VAND 3.0 Challenge
  7. 2026What's the Catch? Evaluating Temporal Consistency in Vision-Language Models
  8. 2025Detect, Classify, Act: Categorizing Industrial Anomalies with Multi-Modal Large Language Models