# Trade-offs in Cross-Domain Generalization of Foundation Model Fine-Tuned for Biometric Applications

arXiv: https://arxiv.org/abs/2509.14921 · HTML: https://mlsanity.com/paper/2509.14921/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 36 of its 70 references that are in the map.
- First posted on arXiv 18 Sep 2025.

## Builds on (most cited first, 30 of 36)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [On the Opportunities and Risks of Foundation Models](https://mlsanity.com/paper/2108.07258/) (2021)
- [Describing Textures in the Wild](https://mlsanity.com/paper/1311.3618/) (2013)
- [Fine-Grained Visual Classification of Aircraft](https://mlsanity.com/paper/1306.5151/) (2013)
- [Do ImageNet Classifiers Generalize to ImageNet?](https://mlsanity.com/paper/1902.10811/) (2019)
- [Reproducible scaling laws for contrastive language-image learning](https://mlsanity.com/paper/2212.07143/) (2022)
- [CosFace: Large Margin Cosine Loss for Deep Face Recognition](https://mlsanity.com/paper/1801.09414/) (2018)
- [Intrinsic Dimensionality Explains the Effectiveness of Language Model Fine-Tuning](https://mlsanity.com/paper/2012.13255/) (2020)
- [An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning](https://mlsanity.com/paper/2308.08747/) (2023)
- [Diffusion Autoencoders: Toward a Meaningful and Decodable Representation](https://mlsanity.com/paper/2111.15640/) (2021)
- [Cross-Age LFW: A Database for Studying Cross-Age Face Recognition in Unconstrained Environments](https://mlsanity.com/paper/1708.08197/) (2017)
- [YFCC100M: The New Data in Multimedia Research](https://mlsanity.com/paper/1503.01817/) (2015)
- [WebFace260M: A Benchmark Unveiling the Power of Million-Scale Deep Face Recognition](https://mlsanity.com/paper/2103.04098/) (2021)
- [ElasticFace: Elastic Margin Loss for Deep Face Recognition](https://mlsanity.com/paper/2109.09416/) (2021)
- [CelebA-Spoof: Large-Scale Face Anti-Spoofing Dataset with Rich Annotations](https://mlsanity.com/paper/2007.12342/) (2020)
- [Arc2Face: A Foundation Model for ID-Consistent Human Faces](https://mlsanity.com/paper/2403.11641/) (2024)
- [Revisiting Catastrophic Forgetting in Large Language Model Tuning](https://mlsanity.com/paper/2406.04836/) (2024)
- [Fine-tuning can cripple your foundation model; preserving features may be the solution](https://mlsanity.com/paper/2308.13320/) (2023)
- [Privacy-friendly Synthetic Data for the Development of Face Morphing Attack Detectors](https://mlsanity.com/paper/2203.06691/) (2022)
- [MorDIFF: Recognition Vulnerability and Attack Detectability of Face Morphing Attacks Created by Diffusion Autoencoders](https://mlsanity.com/paper/2302.01843/) (2023)
- [Synthetic Data for Face Recognition: Current State and Future Prospects](https://mlsanity.com/paper/2305.01021/) (2023)
- [How Good is ChatGPT at Face Biometrics? A First Look into Recognition, Soft Biometrics, and Explainability](https://mlsanity.com/paper/2401.13641/) (2024)
- [MAAD-Face: A Massively Annotated Attribute Dataset for Face Images](https://mlsanity.com/paper/2012.01030/) (2020)

## Abstract (arXiv)

Foundation models such as CLIP have demonstrated exceptional zero- and few-shot transfer capabilities across diverse vision tasks. However, when fine-tuned for highly specialized biometric tasks, face recognition (FR), morphing attack detection (MAD), and presentation attack detection (PAD), these models may suffer from over-specialization. Thus, they may lose one of their foundational strengths, cross-domain generalization. In this work, we systematically quantify these trade-offs by evaluating three instances of CLIP fine-tuned for FR, MAD, and PAD. We evaluate each adapted model as well as the original CLIP baseline on 14 general vision datasets under zero-shot and linear-probe protocols, alongside common FR, MAD, and PAD benchmarks. Our results indicate that fine-tuned models suffer from over-specialization, especially when fine-tuned for complex tasks of FR. Also, our results pointed out that task complexity and classification head design, multi-class (FR) vs. binary (MAD and PAD), correlate with the degree of catastrophic forgetting. The FRoundation model with the ViT-L backbone outperforms other approaches on the large-scale FR benchmark IJB-C, achieving an improvement of up to 58.52%. However, it experiences a substantial performance drop on ImageNetV2, reaching only 51.63% compared to 69.84% achieved by the baseline CLIP model. Moreover, the larger CLIP architecture consistently preserves more of the model's original generalization ability than the smaller variant, indicating that increased model capacity may help mitigate over-specialization.