Tipsomaly (ICASSP 2026)

Official PyTorch implementation of TIPS Over Tricks: Simple Prompts for Effective Zero-shot Anomaly Detection — a spatially-aware zero-shot anomaly detection pipeline built on the TIPS vision-language model, using decoupled prompts and local evidence injection to improve image-level and pixel-level performance.

---

Table of Contents

• 📖 Introduction
• 📊 Results
• 🚀 Quickstart
• 📍 Checkpoints
• 🔧 Setup
• 🗂️ Datasets
• 🛠️ Training
• ⚖️ License
• 🔗 Citation

---

📖 Introduction

Anomaly detection identifies departures from expected behavior in safety-critical settings. When target-domain normal data are unavailable, zero-shot anomaly detection (ZSAD) leverages vision-language models (VLMs). However, CLIP's coarse image-text alignment limits both localization and detection due to (i) spatial misalignment and (ii) weak sensitivity to fine-grained anomalies; prior works compensate with complex auxiliary modules yet largely overlook the choice of backbone. We revisit the backbone and use TIPS-a VLM trained with spatially aware objectives. While TIPS alleviates CLIP's issues, it exposes a distributional gap between global and local features. We address this with decoupled prompts-fixed for image-level detection and learnable for pixel-level localization-and by injecting local evidence into the global score. Without CLIP-specific tricks, our TIPS-based pipeline improves image-level performance by 1.1-3.9% and pixel-level by 1.5-6.9% across seven industrial datasets, delivering strong generalization with a lean architecture.

---

📊 Results

Across 14 industrial and medical benchmarks, Tipsomaly consistently outperforms prior CLIP-based zero-shot methods while remaining lightweight. The figure below summarizes its performance across datasets.

We compare pixel-level anomaly maps from Tipsomaly with prior CLIP-based methods (AdaCLIP and AnomalyCLIP) on industrial and medical samples. As shown in the following figure, Tipsomaly more accurately localizes anomalous regions across both domains.

---

🚀 Quickstart

Quick start (recommended):
Use our ready-to-run Kaggle notebook to reproduce the results with the provided checkpoints.

📍 Checkpoints

You can find our checkpoints trained on MVTec and VisA in the directory workspaces or on google drive at this link.

---

🔧 Setup

To setup the environment and run experiments, follow these steps.

1. Install the required dependencies with the following command.

pip install -r requirements.txt

2. Download your desired checkpoints of the TIPS model. For this purpose, edit and run the mentioned script to download model components. After downloading the components, place them in a desired directory.

bash model/tips/checkpoints/download_checkpoints.sh

mkdir tips && mv tips_oss_l14_highres_distilled_text.npz tips_oss_l14_highres_distilled_vision.npz tokenizer.model tips/

3. Prepare your datasets in the format described in the Datasets section and place them in a desired directory.

mkdir data-root && mv mvtec visa data-root/

4. Now, by setting the correct paths for the model, checkpoints, and the data root path, you can edit the runnable script reproduce.sh to run experiments. Here is a sample:

models_dir="path/to/tips"
data_root_dir="path/to/data-root"
model_version='l14h' 
checkpoint_path="./workspaces/trained_on_mvtec_default/vegan-arkansas/checkpoints"

python test.py --models_dir $models_dir --checkpoint_path $checkpoint_path --data_root_dir $data_root_dir --dataset visa --epoch 2 --model_version $model_version --fixed_prompt_type industrial

🗂️ Datasets

The datasets should be set in the following format. Each dataset, must contain a file named meta.json which contains information about the dataset. Common datasets for anomaly detection and instructions to generate meta.json files can be found in the AdaCLIP GitHub repository. The general expected structure of the datasets is like the following:

data-root/
├── mvtec/
│   ├── meta.json
│   ├── bottle/
│   │   ├── train/
│   │   │   └── good/
│   │   │       └── 000.png
│   │   ├── test/
│   │   │   ├── good/
│   │   │   │   └── 000.png
│   │   │   ├── broken_large/
│   │   │   │   └── 000.png
│   │   │   └── broken_small/
│   │   │       └── 000.png
│   │   └── ground_truth/
│   │       ├── broken_large/
│   │       │   └── 000_mask.png
│   │       └── broken_small/
│   │           └── 000_mask.png
│
├── visa/
├── mpdd/
├── ....
....

🛠️ Training

You can use our ready-to-run Kaggle notebook to train your model and test on your desired dataset. Steps to prepare the datasets and downloading the base model components can be found at the Setup section. You can also find the commands to train the model and test multiple datasets in a loop in the train_test.sh

🔗 Citation

If you find this project helpful for your research, please consider citing the following BibTeX entry.

BibTeX:

@article{salehi2026tips,
  title={TIPS Over Tricks: Simple Prompts for Effective Zero-shot Anomaly Detection},
  author={Salehi, Alireza and Karami, Ehsan and Noey, Sepehr and Noey, Sahand and Yamada, Makoto and Hosseini, Reshad and Sabokrou, Mohammad},
  journal={arXiv preprint arXiv:2602.03594},
  year={2026}
}

⚖️ License

This project is licensed under the MIT License. Please refer to the LICENSE file for more information.