# DIME-FM: DIstilling Multimodal and Efficient Foundation Models

arXiv: https://arxiv.org/abs/2303.18232 · HTML: https://mlsanity.com/paper/2303.18232/

## Key facts

- Cited by 28 papers in the map (data through 14 Oct 2026).
- Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 76 of its 112 references that are in the map.
- First posted on arXiv 31 Mar 2023.

## Placed in

- Multimodal LLMs / vision-language models / [VLM benchmarks and visual reasoning](https://mlsanity.com/mllm/vlm-benchmarks-and-visual-reasoning/)

## Builds on (most cited first, 30 of 76)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://mlsanity.com/paper/2102.05918/) (2021)
- [SSD: Single Shot MultiBox Detector](https://mlsanity.com/paper/1512.02325/) (2015)
- [Unsupervised Learning of Visual Features by Contrasting Cluster Assignments](https://mlsanity.com/paper/2006.09882/) (2020)
- [VQA: Visual Question Answering](https://mlsanity.com/paper/1505.00468/) (2015)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks](https://mlsanity.com/paper/1908.02265/) (2019)
- [Describing Textures in the Wild](https://mlsanity.com/paper/1311.3618/) (2013)
- [The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization](https://mlsanity.com/paper/2006.16241/) (2020)
- [Fine-Grained Visual Classification of Aircraft](https://mlsanity.com/paper/1306.5151/) (2013)
- [FitNets: Hints for Thin Deep Nets](https://mlsanity.com/paper/1412.6550/) (2014)
- [Align before Fuse: Vision and Language Representation Learning with Momentum Distillation](https://mlsanity.com/paper/2107.07651/) (2021)
- [LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs](https://mlsanity.com/paper/2111.02114/) (2021)
- [Do ImageNet Classifiers Generalize to ImageNet?](https://mlsanity.com/paper/1902.10811/) (2019)
- [EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification](https://mlsanity.com/paper/1709.00029/) (2017)

## Built on by (most cited first, 28 of 28)

- [SAM-CLIP: Merging Vision Foundation Models towards Semantic and Spatial Understanding](https://mlsanity.com/paper/2310.15308/) (2023)
- [MobileCLIP: Fast Image-Text Models through Multi-Modal Reinforced Training](https://mlsanity.com/paper/2311.17049/) (2023)
- [A Survey of Resource-efficient LLM and Multimodal Foundation Models](https://mlsanity.com/paper/2401.08092/) (2024)
- [Vision-Language Models for Edge Networks: A Comprehensive Survey](https://mlsanity.com/paper/2502.07855/) (2025)
- [EdgeFM: Leveraging Foundation Model for Open-set Learning on the Edge](https://mlsanity.com/paper/2311.10986/) (2023)
- [CLIP-CID: Efficient CLIP Distillation via Cluster-Instance Discrimination](https://mlsanity.com/paper/2408.09441/) (2024)
- [Knowledge Transfer from Vision Foundation Models for Efficient Training of Small Task-specific Models](https://mlsanity.com/paper/2311.18237/) (2023)
- [PracticalDG: Perturbation Distillation on Vision-Language Models for Hybrid Domain Generalization](https://mlsanity.com/paper/2404.09011/) (2024)
- [Swiss Army Knife: Synergizing Biases in Knowledge from Vision Foundation Models for Multi-Task Learning](https://mlsanity.com/paper/2410.14633/) (2024)
- [LoRKD: Low-Rank Knowledge Decomposition for Medical Foundation Models](https://mlsanity.com/paper/2409.19540/) (2024)
- [Deploying Foundation Model Powered Agent Services: A Survey](https://mlsanity.com/paper/2412.13437/) (2024)
- [Multimodal Unsupervised Domain Generalization by Retrieving Across the Modality Gap](https://mlsanity.com/paper/2402.04416/) (2024)
- [Foundation Models for Structural Health Monitoring](https://mlsanity.com/paper/2404.02944/) (2024)
- [Multimodal Medical Image Binding via Shared Text Embeddings](https://mlsanity.com/paper/2506.18072/) (2025)
- [SynCDR : Training Cross Domain Retrieval Models with Synthetic Data](https://mlsanity.com/paper/2401.00420/) (2023)
- [Exploring the Spectrum of Visio-Linguistic Compositionality and Recognition](https://mlsanity.com/paper/2406.09388/) (2024)
- [Class Incremental Learning with Task-Specific Batch Normalization and Out-of-Distribution Detection](https://mlsanity.com/paper/2411.00430/) (2024)
- [CLAMP: Contrastive LAnguage Model Prompt-tuning](https://mlsanity.com/paper/2312.01629/) (2023)
- [Wisdom of Committee: Diverse Distillation from Large Foundation Models and Domain Experts](https://mlsanity.com/paper/2402.14035/) (2024)
- [YOLO-Vehicle-Pro: A Cloud-Edge Collaborative Framework for Object Detection in Autonomous Driving under Adverse Weather Conditions](https://mlsanity.com/paper/2410.17734/) (2024)
- [Simple yet Effective Semi-supervised Knowledge Distillation from Vision-Language Models via Dual-Head Optimization](https://mlsanity.com/paper/2505.07675/) (2025)
- [ReGATE: Learning Faster and Better with Fewer Tokens in MLLMs](https://mlsanity.com/paper/2507.21420/) (2025)
- [CAST: Contrastive Adaptation and Distillation for Semi-Supervised Instance Segmentation](https://mlsanity.com/paper/2505.21904/) (2025)
- [PCoreSet: Effective Active Learning through Knowledge Distillation from Vision-Language Models](https://mlsanity.com/paper/2506.00910/) (2025)
- [From Semantics, Scene to Instance-awareness: Distilling Foundation Model for Grounded Open-vocabulary Situation Recognition](https://mlsanity.com/paper/2507.14686/) (2025)
- [Make me an Expert: Distilling from Generalist Black-Box Models into Specialized Models for Semantic Segmentation](https://mlsanity.com/paper/2509.00509/) (2025)
- [DynaNav: Dynamic Feature and Layer Selection for Efficient Visual Navigation](https://mlsanity.com/paper/2509.21930/) (2025)
- [Training a Student Expert via Semi-Supervised Foundation Model Distillation](https://mlsanity.com/paper/2604.03841/) (2026)

## Abstract (arXiv)

Large Vision-Language Foundation Models (VLFM), such as CLIP, ALIGN and Florence, are trained on large-scale datasets of image-caption pairs and achieve superior transferability and robustness on downstream tasks, but they are difficult to use in many practical applications due to their large size, high latency and fixed architectures. Unfortunately, recent work shows training a small custom VLFM for resource-limited applications is currently very difficult using public and smaller-scale data. In this paper, we introduce a new distillation mechanism (DIME-FM) that allows us to transfer the knowledge contained in large VLFMs to smaller, customized foundation models using a relatively small amount of inexpensive, unpaired images and sentences. We transfer the knowledge from the pre-trained CLIP-ViTL/14 model to a ViT-B/32 model, with only 40M public images and 28.4M unpaired public sentences. The resulting model "Distill-ViT-B/32" rivals the CLIP-ViT-B/32 model pre-trained on its private WiT dataset (400M image-text pairs): Distill-ViT-B/32 achieves similar results in terms of zero-shot and linear-probing performance on both ImageNet and the ELEVATER (20 image classification tasks) benchmarks. It also displays comparable robustness when evaluated on five datasets with natural distribution shifts from ImageNet.