Multi-View Slot Attention Using Paraphrased Texts for Face Anti-Spoofing
On the map: Multimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
Abstract (arXiv)
Recent face anti-spoofing (FAS) methods have shown remarkable cross-domain performance by employing vision-language models like CLIP. However, existing CLIP-based FAS models do not fully exploit CLIP's patch embedding tokens, failing to detect critical spoofing clues. Moreover, these models rely on a single text prompt per class (e.g., 'live' or 'fake'), which limits generalization. To address these issues, we propose MVP-FAS, a novel framework incorporating two key modules: Multi-View Slot attention (MVS) and Multi-Text Patch Alignment (MTPA). Both modules utilize multiple paraphrased texts to generate generalized features and reduce dependence on domain-specific text. MVS extracts local detailed spatial features and global context from patch embeddings by leveraging diverse texts with multiple perspectives. MTPA aligns patches with multiple text representations to improve semantic robustness. Extensive experiments demonstrate that MVP-FAS achieves superior generalization performance, outperforming previous state-of-the-art methods on cross-domain datasets. Code: https://github.com/Elune001/MVP-FAS.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 41 of its 70 references that are in the map. First posted on arXiv 8 Sep 2025.
Builds on
41 of 70 references.
References · 41
- 2014Adam: A Method for Stochastic Optimizationcited by 37,240
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
- 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
- 2023GPT-4 Technical Reportcited by 18,396
- 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,057
- 2021Learning to Prompt for Vision-Language Modelscited by 2,281
- 2020Object-Centric Learning with Slot Attentioncited by 876
- 2021Generic Attention-model Explainability for Interpreting Bi-Modal and Encoder-Decoder Transformerscited by 320
- 2022ZegCLIP: Towards Adapting CLIP for Zero-shot Semantic Segmentationcited by 178
- 2020Searching Central Difference Convolutional Networks for Face Anti-Spoofingcited by 173
- 2021Self-supervised Video Object Segmentation by Motion Groupingcited by 147
- 2020Single-Side Domain Generalization for Face Anti-Spoofingcited by 139
29 more
- 2022Open Vocabulary Semantic Segmentation with Patch Aligned Contrastive Learningcited by 115
- 2020NAS-FAS: Static-Dynamic Central Difference Network Search for Face Anti-Spoofingcited by 105
- 2023Learning Open-vocabulary Semantic Segmentation Models From Natural Language Supervisioncited by 99
- 2022Object Scene Representation Transformercited by 90
- 2023CLIP the Gap: A Single Domain Generalization Approach for Object Detectioncited by 87
- 2015face anti-spoofing based on color texture analysiscited by 82
- 2022Domain Generalization via Shuffled Style Assembly for Face Anti-Spoofingcited by 73
- 2022PatchNet: A Simple Face Anti-Spoofing Framework via Fine-Grained Patch Recognitioncited by 67
- 2020CASIA-SURF CeFA: A Benchmark for Multi-modal Cross-ethnicity Face Anti-spoofingcited by 66
- 2019CASIA-SURF: A Large-scale Multi-modal Benchmark for Face Anti-spoofingcited by 61
- 2021Self-Domain Adaptation for Face Anti-Spoofingcited by 49
- 2021Adaptive Normalized Representation Learning for Generalizable Face Anti-Spoofingcited by 49
- 2023Rethinking Domain Generalization for Face Anti-spoofing: Separability and Alignmentcited by 48
- 2021Dual Reweighting Domain Generalization for Face Presentation Attack Detectioncited by 47
- 2023FLIP: Cross-domain Face Anti-spoofing with Language Guidancecited by 45
- 2024Improving fine-grained understanding in image-text pre-trainingcited by 44
- 2023Instance-Aware Domain Generalization for Face Anti-Spoofingcited by 42
- 2024CFPL-FAS: Class Free Prompt Learning for Generalizable Face Anti-spoofingcited by 39
- 2022Adaptive Transformers for Robust Few-shot Cross-domain Face Anti-spoofingcited by 38
- 2024Gradient Alignment for Cross-Domain Face Anti-Spoofingcited by 27
- 2023GeoVLN: Learning Geometry-Enhanced Visual Representation with Slot Attention for Vision-and-Language Navigationcited by 24
- 2024Test-Time Domain Generalization for Face Anti-Spoofingcited by 20
- 2024Unified Physical-Digital Face Attack Detectioncited by 17
- 2024Suppress and Rebalance: Towards Generalized Multi-Modal Face Anti-Spoofingcited by 16
- 2023Unsupervised Open-Vocabulary Object Localization in Videoscited by 13
- 2023Compound Text-Guided Prompt Tuning via Image-Adaptive Cuescited by 7
- 2024DiffFAS: Face Anti-Spoofing via Generative Diffusion Modelscited by 7
- 2025Mixture-of-Attack-Experts with Class Regularization for Unified Physical-Digital Face Attack Detectioncited by 5
- 2025FA^{3}-CLIP: Frequency-Aware Cues Fusion and Attack-Agnostic Prompt Learning for Unified Face Attack Detectioncited by 3
Built on by
Citing papers per year: 2026: 1.
1 paper.
Similar, not cited
Close in wording, no citation either way
- 2026Benchmarking Vision Foundation Models for Domain-Generalizable Face Anti-Spoofing
- 2023TeG-DG: Textually Guided Domain Generalization for Face Anti-Spoofing
- 2025FaceShield: Explainable Face Anti-Spoofing with Multimodal Large Language Models
- 2025InstructFLIP: Exploring Unified Vision-Language Model for Face Anti-spoofing
- 2025Interpretable Face Anti-Spoofing: Enhancing Generalization with Multimodal Large Language Models
- 2024SHIELD : An Evaluation Benchmark for Face Spoofing and Forgery Detection with Multimodal Large Language Models
- 2025Harnessing Chain-of-Thought Reasoning in Multimodal Large Language Models for Face Anti-Spoofing
- 2025Learning Unknown Spoof Prompts for Generalized Face Anti-Spoofing Using Only Real Face Images