Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].

One-for-All Few-Shot Anomaly Detection via Instance-Induced Prompt Learning

Authors: Wenxi Lv, Qinliang Su, Wenchao Xu

ICLR 2025 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experimental results on MVTec and Vis A demonstrate the superiority of our method in few-shot anomaly detection task under the one-for-all paradigm. Our code is available in https://github.com/Vanssssry/One-For-All-Few-Shot-Anomaly-Detection.
Researcher Affiliation Academia 1 School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China. 2 Guangdong Key Laboratory of Big Data Analysis and Processing, Guangzhou, China. 3 Department of Computing, The Hong Kong Polytechnic University, Hong Kong SAR
Pseudocode No The paper describes the methodology using textual explanations and a framework diagram (Figure 1), but does not include any explicitly labeled pseudocode or algorithm blocks.
Open Source Code Yes Our code is available in https://github.com/Vanssssry/One-For-All-Few-Shot-Anomaly-Detection.
Open Datasets Yes In this paper, we conduct extensive experiments on benchmarks MVTec (Bergmann et al., 2019) and Vis A (Zou et al., 2022).
Dataset Splits Yes We conduct the comparison experiments between our method and the latest methods under the setting that the training dataset contains all categories of the benchmark and only K-shot normal image for each class. Each class contains at most K examples, where K typically ranges from 1 to 4.
Hardware Specification Yes All experiments are conducted with a single NVIDIA RTX 3090 24GB GPU.
Software Dependencies No The paper mentions using 'CLIP (VIT-L-14)', 'BLIP-Diffusion' for Q-Former, and 'Py Torch' for implementation, but does not provide specific version numbers for these software components.
Experiment Setup Yes The length of learnable prompt T is set as 24. The length of compressed category tokens m is 4. The projection networks used to incorporate visual information are two-layer neural network. The weight factors α, β that balance loss are both fixed as 8. ... We use the Adam optimizer (Kingma & Ba, 2015) with a learning rate of 0.001 to update model parameters. The epoch is 20 for all experiments.