Update dataset card

README.md

@@ -1,48 +1,101 @@
 ---
+license: cc-by-4.0
+language:
+- en
 pretty_name: VisAnomBench
+size_categories:
+- 1K<n<10K
+task_categories:
+- image-text-to-text
 tags:
-- anomaly-detection
 - time-series
+- anomaly-detection
 - vision-language
+- explanations
 ---
 
 # VisAnomBench
 
-VisAnomBench is a benchmark for vision-language reasoning over time-series anomaly plots. It is designed to evaluate whether a model can inspect a plotted time series, decide if an anomaly is present, localize the anomalous interval, and explain the visual evidence behind the decision.
+**Tiny but Trusted: Efficient Vision-Language Reasoning for Time-Series Anomaly Detection**
+
+**Xiaona Zhou, Muntasir Wahed, Tianjiao Yu, Constantin Brif, Ismini Lourentzou**
+
+### 🎉 Accepted to NeurIPS 2026
+
+[📄 Paper](https://arxiv.org/abs/2605.30344) · [🌐 Project Page](https://plan-lab.github.io/projects/VisAnom/) · [💻 Code](https://github.com/PLAN-Lab/VisAnom) · [🤗 VisAnomReasoner-3B](https://huggingface.co/PLAN-Lab/VisAnomReasoner-3B) · [🤗 VisAnomReasoner-7B](https://huggingface.co/PLAN-Lab/VisAnomReasoner-7B)
+
+---
+
+## Overview
+
+VisAnomBench is an explanation-augmented benchmark for vision-language time-series anomaly reasoning. Each sample pairs a rendered time-series plot with a prompt and ground-truth anomaly intervals. Training samples also include a step-by-step explanation grounded in the plot. The benchmark is used to train and evaluate [VisAnomReasoner-3B](https://huggingface.co/PLAN-Lab/VisAnomReasoner-3B) and [VisAnomReasoner-7B](https://huggingface.co/PLAN-Lab/VisAnomReasoner-7B).
+
+<p align="center">
+  <img src="VisAnomReasoner_teaser.png" width="100%">
+</p>
+
+Given a time-series plot (left top), VisAnomReasoner, trained on VisAnomBench, locates anomalies while providing details grounded in the plot (left bottom). On the right, VisAnomReasoner-7B outperforms the strongest baselines on VisAnomBench and TSB-AD-U.
+
+---
+
+## Dataset Structure
+
+| Split | Samples | UCR-EGI | GutenTAG | UCR-TSAD | KPI |
+|---|---:|---:|---:|---:|---:|
+| train | 2,576 | 1,682 | 563 | 200 | 131 |
+| test | 740 | 415 | 247 | 49 | 29 |
+
+### Data Fields
+
+| Field | Description |
+|---|---|
+| `image` | Path to the plot image, relative to the root of the [VisAnom code repository](https://github.com/PLAN-Lab/VisAnom), for example `VisAnomBench/VisAnomBench_plot_test/test_0.png`. |
+| `query` | Prompt given to the model: context about the series, task instructions, and the required output format. |
+| `resp` | Target response: image description, rationale, step-by-step reasoning, and a final answer of the form `<anomaly>True</anomaly><index>(start, end)</index>` followed by the reasoning in `<think>...</think>`. Empty in the test split. |
+| `anomalies` | Ground-truth anomaly intervals as inclusive `(start, end)` pairs, for example `(52000, 52620)`. KPI intervals use timestamps, for example `(2016-11-09 03:35:00, 2016-11-09 03:45:00)`. |
+
+## Download
 
-Unlike many time-series anomaly benchmarks that mainly provide raw numeric sequences and anomaly labels, VisAnomBench pairs rendered plots with language prompts. The training split also includes reference answers that localize anomalies and explain the visual evidence, making it suitable for supervised fine-tuning (SFT) of vision-language models.
+To use the data with the VisAnom code, run this from the root of the code repository:
 
-## Contents
+```bash
+hf download PLAN-Lab/VisAnomBench \
+  --repo-type dataset \
+  --include "VisAnomBench_train.json" \
+  --include "VisAnomBench_test.json" \
+  --include "VisAnomBench_plot_train/*" \
+  --include "VisAnomBench_plot_test/*" \
+  --include "VisAnomBench_train_ts/*" \
+  --include "VisAnomBench_test_ts/*" \
+  --local-dir VisAnomBench
+```
+
+## How the Explanations Were Created
+
+For each time-series segment, four general-purpose VLMs (Grok-4-Fast, LLaMA-4-Maverick, Gemma-3-27B-IT, and Qwen2.5-VL-32B) generated candidate outputs, each with an anomaly decision, predicted intervals, and a reasoning trace. Candidates were ranked by a composite reward that combines anomaly localization accuracy with explanation quality (visual groundedness, axis awareness, and clarity) scored by Qwen2.5-VL-72B, and the highest-scoring candidate was kept as the target response. See the paper for details.
 
-| Path | What it contains |
-| --- | --- |
-| `VisAnomBench_train.json` | Training prompts, image paths, anomaly intervals, and reference answers with visual explanations for SFT. |
-| `VisAnomBench_test.json` | Test prompts, image paths, and anomaly intervals for evaluation. Reference explanation responses are not included. |
-| `VisAnomBench_plot_train/` | Rendered plot images used by the training JSON. |
-| `VisAnomBench_plot_test/` | Rendered plot images used by the test JSON. |
-| `VisAnomBench_train_ts/` | Raw time-series CSV files for the training split. |
-| `VisAnomBench_test_ts/` | Raw time-series CSV files for the test split. |
+## Source Datasets
 
-## Split Sizes
+| Source | Reference | License |
+|---|---|---|
+| [KPI](https://github.com/NetManAIOps/KPI-Anomaly-Detection) | Xu et al., "Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applications," WWW 2018 | MIT |
+| [GutenTAG](https://github.com/TimeEval/GutenTAG) | Wenig et al., "TimeEval: A Benchmarking Toolkit for Time Series Anomaly Detection Algorithms," PVLDB 2022 | MIT |
+| [UCR-EGI](https://doi.org/10.5441/002/edbt.2020.09) | Gao, Lin, and Brif, "Ensemble Grammar Induction for Detecting Anomalies in Time Series," EDBT 2020 | See source |
+| [UCR-TSAD](https://www.cs.ucr.edu/~eamonn/time_series_data_2018/) | Wu and Keogh, "Current Time Series Anomaly Detection Benchmarks are Flawed and are Creating the Illusion of Progress," TKDE 2021 | See source |
 
-| Split | Time series |
-| --- | ---: |
-| Train | 2,576 |
-| Test | 740 |
+The source time series remain subject to the terms of their original sources. Please also cite the source datasets when using VisAnomBench.
 
-Each JSON record contains an `image` path and a `query` prompt. Training records also include a `resp` field with the reference answer used for SFT. The `anomalies` field stores the ground-truth interval for scoring. CSV files contain `timestamp`, `value`, and `label` columns.
+## License
+
+VisAnomBench is released under the [CC BY 4.0 license](https://creativecommons.org/licenses/by/4.0/).
 
 ## Citation
 
 ```bibtex
-@article{zhou2026tiny,
+@inproceedings{zhou2026tiny,
   title={Tiny but Trusted: Efficient Vision-Language Reasoning for Time-Series Anomaly Detection},
   author={Zhou, Xiaona and Wahed, Muntasir and Yu, Tianjiao and Brif, Constantin and Lourentzou, Ismini},
-  journal={arXiv preprint arXiv:2605.30344},
+  booktitle={Advances in Neural Information Processing Systems (NeurIPS)},
   year={2026}
 }
 ```
-
-## Contact
-
-For questions or suggestions, contact Xiaona Zhou at xiaonaz2@illinois.edu.