# Captured by Captions: On Memorization and its Mitigation in CLIP Models

arXiv: https://arxiv.org/abs/2502.07830 · HTML: https://mlsanity.com/paper/2502.07830/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 39 references that are in the map.
- First posted on arXiv 11 Feb 2025.

## Builds on (most cited first, 29 of 29)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [Extracting Training Data from Large Language Models](https://mlsanity.com/paper/2012.07805/) (2020)
- [Understanding deep learning requires rethinking generalization](https://mlsanity.com/paper/1611.03530/) (2016)
- [Reproducible scaling laws for contrastive language-image learning](https://mlsanity.com/paper/2212.07143/) (2022)
- [A Closer Look at Memorization in Deep Networks](https://mlsanity.com/paper/1706.05394/) (2017)
- [The Secret Sharer: Evaluating and Testing Unintended Memorization in Neural Networks](https://mlsanity.com/paper/1802.08232/) (2018)
- [Benign Overfitting in Linear Regression](https://mlsanity.com/paper/1906.11300/) (2019)
- [What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimation](https://mlsanity.com/paper/2008.03703/) (2020)
- [Does Learning Require Memorization? A Short Tale about a Long Tail](https://mlsanity.com/paper/1906.05271/) (2019)
- [Tip-Adapter: Training-free Adaption of CLIP for Few-shot Classification](https://mlsanity.com/paper/2207.09519/) (2022)
- [Memorization Without Overfitting: Analyzing the Training Dynamics of Large Language Models](https://mlsanity.com/paper/2205.10770/) (2022)
- [Improving CLIP Training with Language Rewrites](https://mlsanity.com/paper/2305.20088/) (2023)
- [Machine Learning Models that Remember Too Much](https://mlsanity.com/paper/1709.07886/) (2017)
- [The Privacy Onion Effect: Memorization is Relative](https://mlsanity.com/paper/2206.10469/) (2022)
- [Measuring Forgetting of Memorized Training Examples](https://mlsanity.com/paper/2207.00099/) (2022)
- [CLIP Models are Few-shot Learners: Empirical Studies on VQA and Visual Entailment](https://mlsanity.com/paper/2203.07190/) (2022)
- [Generalization v.s. Memorization: Tracing Language Models' Capabilities Back to Pretraining Data](https://mlsanity.com/paper/2407.14985/) (2024)
- [EncoderMI: Membership Inference against Pre-trained Encoders in Contrastive Learning](https://mlsanity.com/paper/2108.11023/) (2021)
- [Improving Zero-Shot Generalization for CLIP with Synthesized Prompts](https://mlsanity.com/paper/2307.07397/) (2023)
- [Retrieving-to-Answer: Zero-Shot Video Question Answering with Frozen Large Language Models](https://mlsanity.com/paper/2306.11732/) (2023)
- [Do SSL Models Have D'ej`a Vu? A Case of Unintended Memorization in Self-supervised Learning](https://mlsanity.com/paper/2304.13850/) (2023)
- [Memorization in Self-Supervised Learning Improves Downstream Generalization](https://mlsanity.com/paper/2401.12233/) (2024)
- [Normalizing Flow based Feature Synthesis for Outlier-Aware Object Detection](https://mlsanity.com/paper/2302.07106/) (2023)
- [CLIP-Decoder : ZeroShot Multilabel Classification using Multimodal CLIP Aligned Representation](https://mlsanity.com/paper/2406.14830/) (2024)
- [Localizing Memorization in SSL Vision Encoders](https://mlsanity.com/paper/2409.19069/) (2024)
- [On the Memorization Properties of Contrastive Learning](https://mlsanity.com/paper/2107.10143/) (2021)

## Built on by (most cited first, 4 of 4)

- [LeakyCLIP: Extracting Training Data from CLIP](https://mlsanity.com/paper/2508.00756/) (2025)
- [Demystifying Foreground-Background Memorization in Diffusion Models](https://mlsanity.com/paper/2508.12148/) (2025)
- [Localizing and Mitigating Memorization in Image Autoregressive Models](https://mlsanity.com/paper/2509.00488/) (2025)
- [MultiMem: Measuring and Mitigating Memorization in Multi-Modal Contrastive Learning](https://mlsanity.com/paper/2606.22220/) (2026)

## Abstract (arXiv)

Multi-modal models, such as CLIP, have demonstrated strong performance in aligning visual and textual representations, excelling in tasks like image retrieval and zero-shot classification. Despite this success, the mechanisms by which these models utilize training data, particularly the role of memorization, remain unclear. In uni-modal models, both supervised and self-supervised, memorization has been shown to be essential for generalization. However, it is not well understood how these findings would apply to CLIP, which incorporates elements from both supervised learning via captions that provide a supervisory signal similar to labels, and from self-supervised learning via the contrastive objective. To bridge this gap in understanding, we propose a formal definition of memorization in CLIP (CLIPMem) and use it to quantify memorization in CLIP models. Our results indicate that CLIP's memorization behavior falls between the supervised and self-supervised paradigms, with "mis-captioned" samples exhibiting highest levels of memorization. Additionally, we find that the text encoder contributes more to memorization than the image encoder, suggesting that mitigation strategies should focus on the text domain. Building on these insights, we propose multiple strategies to reduce memorization while at the same time improving utility--something that had not been shown before for traditional learning paradigms where reducing memorization typically results in utility decrease.