# Hyper-FSAD: Training-Free and Language-Free Few-Shot Anomaly Detection via Sparse Hyper Matching

arXiv: https://arxiv.org/abs/2605.10628 · HTML: https://mlsanity.com/paper/2605.10628/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 24 references that are in the map.
- First posted on arXiv 11 May 2026.

## Builds on (most cited first, 9 of 9)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [VT-ADL: A Vision Transformer Network for Image Anomaly Detection and Localization](https://mlsanity.com/paper/2104.10036/) (2021)
- [Sparse Sequence-to-Sequence Models](https://mlsanity.com/paper/1905.05702/) (2019)
- [VCP-CLIP: A visual context prompting model for zero-shot anomaly segmentation](https://mlsanity.com/paper/2407.12276/) (2024)
- [The Impact of Scanner Domain Shift on Deep Learning Performance in Medical Imaging: an Experimental Study](https://mlsanity.com/paper/2409.04368/) (2024)
- [SubspaceAD: Training-Free Few-Shot Anomaly Detection via Subspace Modeling](https://mlsanity.com/paper/2602.23013/) (2026)
- [Investigating Shift Equivalence of Convolutional Neural Networks in Industrial Defect Segmentation](https://mlsanity.com/paper/2309.16902/) (2023)
- [Is Task-Specific Training Necessary for Anomaly Detection?](https://mlsanity.com/paper/2601.22763/) (2026)

## Abstract (arXiv)

Few-shot anomaly detection (FSAD) is particularly valuable when only a few normal images are available in a new target domain, while anomalous cases are rare, diverse, and difficult to enumerate in advance. However, existing methods often still require task-specific fitting or language prompts, and their patch-level retrieval commonly relies on brittle nearest-neighbor or fixed Top-p rules. We propose Hyper-FSAD, a training-free and language-free framework that performs support-only inference with a frozen visual encoder. We formulate FSAD as support-only scoring in frozen feature space and establish the selective stability of sparse retrieval and sufficient conditions for normal--anomaly separation. Guided by this analysis, Sparse Hyper Matching uses textitsparsemax to adaptively select support patches for each query patch, exactly suppressing below-threshold distractors without a manually specified retrieval hyperparameter. Dual-Branch Image Scoring further combines local reconstruction discrepancies with support-conditioned global <CLS> deviation. Across four industrial and two medical benchmarks, Hyper-FSAD achieves the best overall performance across the 1/2/4-shot settings, while requiring only 52.6,ms per image and 0.89,GB GPU memory. The code will be released.