# EXPL-FR: Explaining Face Recognition Models via Vision-Language Alignment

arXiv: https://arxiv.org/abs/2608.21486 · HTML: https://mlsanity.com/paper/2608.21486/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 43 of its 59 references that are in the map.
- First posted on arXiv 21 Aug 2026.

## Builds on (most cited first, 30 of 43)

- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)](https://mlsanity.com/paper/1711.11279/) (2017)
- [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https://mlsanity.com/paper/2502.14786/) (2025)
- [Concept Bottleneck Models](https://mlsanity.com/paper/2007.04612/) (2020)
- [CLIP-Adapter: Better Vision-Language Models with Feature Adapters](https://mlsanity.com/paper/2110.04544/) (2021)
- [MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition](https://mlsanity.com/paper/1607.08221/) (2016)
- [Mind the Gap: Understanding the Modality Gap in Multi-modal Contrastive Representation Learning](https://mlsanity.com/paper/2203.02053/) (2022)
- [Learning Face Representation from Scratch](https://mlsanity.com/paper/1411.7923/) (2014)
- [AdaFace: Quality Adaptive Margin for Face Recognition](https://mlsanity.com/paper/2204.00964/) (2022)
- [Tip-Adapter: Training-free Adaption of CLIP for Few-shot Classification](https://mlsanity.com/paper/2207.09519/) (2022)
- [MagFace: A Universal Representation for Face Recognition and Quality Assessment](https://mlsanity.com/paper/2103.06627/) (2021)
- [Language in a Bottle: Language Model Guided Concept Bottlenecks for Interpretable Image Classification](https://mlsanity.com/paper/2211.11158/) (2022)
- [Post-hoc Concept Bottleneck Models](https://mlsanity.com/paper/2205.15480/) (2022)
- [Cross-Age LFW: A Database for Studying Cross-Age Face Recognition in Unconstrained Environments](https://mlsanity.com/paper/1708.08197/) (2017)
- [Interpreting CLIP's Image Representation via Text-Based Decomposition](https://mlsanity.com/paper/2310.05916/) (2023)
- [CLIP-Dissect: Automatic Description of Neuron Representations in Deep Vision Networks](https://mlsanity.com/paper/2204.10965/) (2022)
- [ElasticFace: Elastic Margin Loss for Deep Face Recognition](https://mlsanity.com/paper/2109.09416/) (2021)
- [GAN-Control: Explicitly Controllable GANs](https://mlsanity.com/paper/2101.02477/) (2021)
- [Low-Resolution Face Recognition](https://mlsanity.com/paper/1811.08965/) (2018)
- [Rosetta Neurons: Mining the Common Units in a Model Zoo](https://mlsanity.com/paper/2306.09346/) (2023)
- [WebFace260M: A Benchmark for Million-Scale Deep Face Recognition](https://mlsanity.com/paper/2204.10149/) (2022)
- [CFPL-FAS: Class Free Prompt Learning for Generalizable Face Anti-spoofing](https://mlsanity.com/paper/2403.14333/) (2024)
- [KeyPoint Relative Position Encoding for Face Recognition](https://mlsanity.com/paper/2403.14852/) (2024)
- [Explainable Face Recognition](https://mlsanity.com/paper/2008.00916/) (2020)
- [How Good is ChatGPT at Face Biometrics? A First Look into Recognition, Soft Biometrics, and Explainability](https://mlsanity.com/paper/2401.13641/) (2024)
- [Face Quality Estimation and Its Correlation to Demographic and Non-Demographic Bias in Face Recognition](https://mlsanity.com/paper/2004.01019/) (2020)
- [xCos: An Explainable Cosine Metric for Face Verification Task](https://mlsanity.com/paper/2003.05383/) (2020)
- [CLIP-PAE: Projection-Augmentation Embedding to Extract Relevant Features for a Disentangled, Interpretable, and Controllable Text-Guided Face Manipulation](https://mlsanity.com/paper/2210.03919/) (2022)
- [FaceLLM: A Multimodal Large Language Model for Face Understanding](https://mlsanity.com/paper/2507.10300/) (2025)

## Abstract (arXiv)

Deep face recognition (FR) models reach near-saturated accuracy but remain opaque: a practitioner cannot ask which semantic attributes a similarity score relied upon. EXPL-FR answers this inside the FR model's own embedding space. A lightweight adapter aligns a vision-language model's (VLM) image encoder with the frozen FR space, trained on face images alone and never on text. Because the VLM's encoders share one space, the same adapter applies to the text encoder, turning 978 attribute prompts in 22 categories, also extendable, into FR-space anchors at no extra cost. We do not assume this transfer works: a face-verification protocol measures it, and an ablation changing only the adapter isolates its contribution. Not every concept survives, because an FR model earns its invariances by discarding the factors it must verify identities across. A label-free detectability measure compares each concept's separability in FR space against the VLM space, and the 100 most detectable form the model's readable semantic signature, which separates identities better than the full vocabulary. We cover four FR backbones and two VLM encoders, EXPL-FR needs no architecture access, and supports identity-level, per-image, and differential explanations. We benchmark attribute-level auditing under three supervision settings, human labels (current practice), VLM pseudo-labels, and our fully prompt-driven audit, against real verification behavior. With no labels, the prompt-driven audit ranks four FR models by their measured per-ethnicity RFW errors and ranks controlled attribute changes by their true verification cost.