Contamination MapVision / Model
AdaptCLIP, trained on VisA
- Organisation
- Tencent YouTu Lab et al.
- Family
- AdaptCLIP · version 12_4_128_train_on_visa_3adapters_batch8/epoch_15.pth
- Released
- 2025-05-15 (arXiv v1 submission date (https://arxiv.org/abs/2505.09926))
- Checkpoints
- csgaobb/AdaptCLIP adaptclip_checkpoints/12_4_128_train_on_visa_3adapters_batch8/epoch_15.pth
- Weights
- Open · gpl-2.0
- Links
- PaperModel cardCode
Training data disclosure: Full list
Level determined for 12% of benchmark cells.
- TestTest data seen1
- SeenSeen, split unclear0
- TrainTraining data seen0
- Related data seen0
- NoNot seen0
- ?Unknown7
“Following AnomalyCLIP [53], we train AdaptCLIP using the testing data from MVTec and evaluate zero-/few-shot performance on other datasets. As for the evaluation of MVTec, we train AdaptCLIP using the testing data of VisA.”
“Trained on VisA dataset”
Built on
Everything these base models may have seen is inherited by AdaptCLIP, trained on VisA.
-
OpenAI CLIP
· disclosure: Partial list
Evidence
“We utilize the pre-trained CLIP (ViT-L/14@336) as the default CLIP model”
arxiv.org · paper · Appendix (implementation details) · retrieved 2026-10-15 “base_model: - openai/clip-vit-large-patch14-336”
huggingface.co · model card · YAML metadata · retrieved 2026-10-15
Other declared training data
-
VisA · split: test · fine-tuning
Evidence
“Following AnomalyCLIP [53], we train AdaptCLIP using the testing data from MVTec and evaluate zero-/few-shot performance on other datasets. As for the evaluation of MVTec, we train AdaptCLIP using the testing data of VisA.”
arxiv.org · paper · Section 4.1, Training and Testing Protocol · retrieved 2026-10-15 “As for the evaluation of MVTec, we train models using VisA's test data.”
arxiv.org · paper · Appendix, datasets · retrieved 2026-10-15
Benchmark cells
Industrial anomaly detection
Unknown MVTec AD · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Unknown MVTec AD 2 · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Unknown MVTec LOCO AD · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Test data seen VisA · in fine-tuning data: its test split was used for training
Evidence (2)
“Following AnomalyCLIP [53], we train AdaptCLIP using the testing data from MVTec and evaluate zero-/few-shot performance on other datasets. As for the evaluation of MVTec, we train AdaptCLIP using the testing data of VisA.”
“As for the evaluation of MVTec, we train models using VisA's test data.”
Unknown Real-IAD · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Unknown MPDD · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Unknown BTAD · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Unknown KolektorSDD2 · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Notes
Used only for MVTec results.
Sources
- AdaptCLIP: Adapting CLIP for Universal Visual Anomaly Detection · paper · retrieved 2026-10-15
- AdaptCLIP README · readme · retrieved 2026-10-15
- csgaobb/AdaptCLIP model card (gated repo; raw URL returns 401, resolve URL used) · model card · retrieved 2026-10-15