# Continual Vision-Language Representation Learning with Off-Diagonal Information

arXiv: https://arxiv.org/abs/2305.07437 · HTML: https://mlsanity.com/paper/2305.07437/

## Key facts

- Cited by 40 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 50 references that are in the map.
- First posted on arXiv 11 May 2023.

## Builds on (most cited first, 30 of 38)

- [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)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [Learning without Forgetting](https://mlsanity.com/paper/1606.09282/) (2016)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [Barlow Twins: Self-Supervised Learning via Redundancy Reduction](https://mlsanity.com/paper/2103.03230/) (2021)
- [LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs](https://mlsanity.com/paper/2111.02114/) (2021)
- [Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Concepts](https://mlsanity.com/paper/2102.08981/) (2021)
- [Experience Replay for Continual Learning](https://mlsanity.com/paper/1811.11682/) (2018)
- [Learning to Prompt for Continual Learning](https://mlsanity.com/paper/2112.08654/) (2021)
- [Dark Experience for General Continual Learning: a Strong, Simple Baseline](https://mlsanity.com/paper/2004.07211/) (2020)
- [Dense Contrastive Learning for Self-Supervised Visual Pre-Training](https://mlsanity.com/paper/2011.09157/) (2020)
- [Supervision Exists Everywhere: A Data Efficient Contrastive Language-Image Pre-training Paradigm](https://mlsanity.com/paper/2110.05208/) (2021)
- [Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Models](https://mlsanity.com/paper/2209.07511/) (2022)
- [DER: Dynamically Expandable Representation for Class Incremental Learning](https://mlsanity.com/paper/2103.16788/) (2021)
- [DetCo: Unsupervised Contrastive Learning for Object Detection](https://mlsanity.com/paper/2102.04803/) (2021)
- [LAMOL: LAnguage MOdeling for Lifelong Language Learning](https://mlsanity.com/paper/1909.03329/) (2019)
- [Continual Lifelong Learning in Natural Language Processing: A Survey](https://mlsanity.com/paper/2012.09823/) (2020)
- [Uncertainty-based Continual Learning with Adaptive Regularization](https://mlsanity.com/paper/1905.11614/) (2019)
- [SS-IL: Separated Softmax for Incremental Learning](https://mlsanity.com/paper/2003.13947/) (2020)
- [Representational Continuity for Unsupervised Continual Learning](https://mlsanity.com/paper/2110.06976/) (2021)
- [MVP: Multimodality-guided Visual Pre-training](https://mlsanity.com/paper/2203.05175/) (2022)
- [SSUL: Semantic Segmentation with Unknown Label for Exemplar-based Class-Incremental Learning](https://mlsanity.com/paper/2106.11562/) (2021)
- [ORDisCo: Effective and Efficient Usage of Incremental Unlabeled Data for Semi-supervised Continual Learning](https://mlsanity.com/paper/2101.00407/) (2021)
- [CLiMB: A Continual Learning Benchmark for Vision-and-Language Tasks](https://mlsanity.com/paper/2206.09059/) (2022)

## Built on by (most cited first, 30 of 40)

- [Continual Learning of Large Language Models: A Comprehensive Survey](https://mlsanity.com/paper/2404.16789/) (2024)
- [Lifelong Learning of Large Language Model based Agents: A Roadmap](https://mlsanity.com/paper/2501.07278/) (2025)
- [Continual Instruction Tuning for Large Multimodal Models](https://mlsanity.com/paper/2311.16206/) (2023)
- [Recent Advances of Foundation Language Models-based Continual Learning: A Survey](https://mlsanity.com/paper/2405.18653/) (2024)
- [Recent Advances of Multimodal Continual Learning: A Comprehensive Survey](https://mlsanity.com/paper/2410.05352/) (2024)
- [TiC-CLIP: Continual Training of CLIP Models](https://mlsanity.com/paper/2310.16226/) (2023)
- [GraphControl: Adding Conditional Control to Universal Graph Pre-trained Models for Graph Domain Transfer Learning](https://mlsanity.com/paper/2310.07365/) (2023)
- [A Unified Approach to Domain Incremental Learning with Memory: Theory and Algorithm](https://mlsanity.com/paper/2310.12244/) (2023)
- [Beyond Anti-Forgetting: Multimodal Continual Instruction Tuning with Positive Forward Transfer](https://mlsanity.com/paper/2401.09181/) (2024)
- [Degeneration-Tuning: Using Scrambled Grid shield Unwanted Concepts from Stable Diffusion](https://mlsanity.com/paper/2308.02552/) (2023)
- [CLAP4CLIP: Continual Learning with Probabilistic Finetuning for Vision-Language Models](https://mlsanity.com/paper/2403.19137/) (2024)
- [Synthetic Data is an Elegant GIFT for Continual Vision-Language Models](https://mlsanity.com/paper/2503.04229/) (2025)
- [Continual Learning for VLMs: A Survey and Taxonomy Beyond Forgetting](https://mlsanity.com/paper/2508.04227/) (2025)
- [LLMs Can Evolve Continually on Modality for X-Modal Reasoning](https://mlsanity.com/paper/2410.20178/) (2024)
- [Continual Learning for Smart City: A Survey](https://mlsanity.com/paper/2404.00983/) (2024)
- [Fine Tuning without Catastrophic Forgetting via Selective Low Rank Adaptation](https://mlsanity.com/paper/2501.15377/) (2025)
- [Mitigating Intra- and Inter-modal Forgetting in Continual Learning of Unified Multimodal Models](https://mlsanity.com/paper/2512.03125/) (2025)
- [Premonition: Using Generative Models to Preempt Future Data Changes in Continual Learning](https://mlsanity.com/paper/2403.07356/) (2024)
- [Continual Retinal Vision-Language Pre-training upon Incremental Imaging Modalities](https://mlsanity.com/paper/2506.19320/) (2025)
- [Affordance-First Decomposition for Continual Learning in Video-Language Understanding](https://mlsanity.com/paper/2512.00694/) (2025)
- [OVMR: Open-Vocabulary Recognition with Multi-Modal References](https://mlsanity.com/paper/2406.04675/) (2024)
- [Branch, or Layer? Zeroth-Order Optimization for Continual Learning of Vision-Language Models](https://mlsanity.com/paper/2506.12409/) (2025)
- [CATA: Continual Machine Unlearning via Conflict-Averse Task Arithmetic](https://mlsanity.com/paper/2605.18610/) (2026)
- [Hyperbolic Multimodal Continual Learning](https://mlsanity.com/paper/2608.09572/) (2026)
- [The Inter-Intra Modal Measure: A Predictive Lens on Fine-Tuning Outcomes in Vision-Language Models](https://mlsanity.com/paper/2407.15731/) (2024)
- [DSCA: Dynamic Subspace Concept Alignment for Lifelong VLM Editing](https://mlsanity.com/paper/2604.07965/) (2026)
- [Beyond Classification: Dynamic Adapter Routing for Continual Multimodal Retrieval](https://mlsanity.com/paper/2605.31229/) (2026)
- [Attention-Spectrum Regularization for Replay-Free Continual Multimodal LLMs](https://mlsanity.com/paper/2606.23063/) (2026)
- [Hidden Forgetting in Continual Multimodal Learning: When Accuracy Survives but Grounding Fails](https://mlsanity.com/paper/2607.02020/) (2026)
- [Continual Contrastive Spoken Language Understanding](https://mlsanity.com/paper/2310.02699/) (2023)

## Abstract (arXiv)

Large-scale multi-modal contrastive learning frameworks like CLIP typically require a large amount of image-text samples for training. However, these samples are always collected continuously in real scenarios. This paper discusses the feasibility of continual CLIP training using streaming data. Unlike continual learning based on self-supervised learning methods for pure images, which is empirically robust against catastrophic forgetting, CLIP's performance degeneration in the continual setting is significant and non-neglectable. By analyzing the changes in the model's representation space during continual CLIP training from a spatial geometry perspective, we explore and summarize these spatial variations as Spatial Disorder (SD), which can be divided into Intra-modal Rotation and Inter-modal Deviation. Moreover, we empirically and theoretically demonstrate how SD leads to a performance decline for CLIP on cross-modal retrieval tasks. To alleviate SD, we propose a new continual vision-language representation learning framework Mod-X: Maintain off-diagonal information-matriX. By selectively aligning the off-diagonal information distribution of contrastive matrices, the Mod-X improves the capability of the multi-modal model by maintaining the multi-modal representation space alignment on the old data domain during continuously fitting the new training data domain. Experiments on commonly used datasets with different scales and scopes have demonstrated the effectiveness of our method.