# Improving fine-grained understanding in image-text pre-training

arXiv: https://arxiv.org/abs/2401.09865 · HTML: https://mlsanity.com/paper/2401.09865/

## Key facts

- Cited by 44 papers in the map (data through 14 Oct 2026).
- Builds on 46 of its 57 references that are in the map.
- First posted on arXiv 18 Jan 2024.

## Builds on (most cited first, 30 of 46)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [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)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://mlsanity.com/paper/2102.05918/) (2021)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [Benchmarking Neural Network Robustness to Common Corruptions and Perturbations](https://mlsanity.com/paper/1903.12261/) (2019)
- [Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations](https://mlsanity.com/paper/1602.07332/) (2016)
- [Survey of Hallucination in Natural Language Generation](https://mlsanity.com/paper/2202.03629/) (2022)
- [SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing](https://mlsanity.com/paper/1808.06226/) (2018)
- [The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization](https://mlsanity.com/paper/2006.16241/) (2020)
- [Align before Fuse: Vision and Language Representation Learning with Momentum Distillation](https://mlsanity.com/paper/2107.07651/) (2021)
- [Do ImageNet Classifiers Generalize to ImageNet?](https://mlsanity.com/paper/1902.10811/) (2019)
- [Natural Adversarial Examples](https://mlsanity.com/paper/1907.07174/) (2019)
- [Flickr30k Entities: Collecting Region-to-Phrase Correspondences for Richer Image-to-Sentence Models](https://mlsanity.com/paper/1505.04870/) (2015)
- [CoCa: Contrastive Captioners are Image-Text Foundation Models](https://mlsanity.com/paper/2205.01917/) (2022)
- [LVIS: A Dataset for Large Vocabulary Instance Segmentation](https://mlsanity.com/paper/1908.03195/) (2019)
- [Grounded Language-Image Pre-training](https://mlsanity.com/paper/2112.03857/) (2021)
- [Learning Robust Global Representations by Penalizing Local Predictive Power](https://mlsanity.com/paper/1905.13549/) (2019)
- [Publicly Available Clinical BERT Embeddings](https://mlsanity.com/paper/1904.03323/) (2019)
- [Scaling Vision Transformers](https://mlsanity.com/paper/2106.04560/) (2021)
- [Revisiting Unreasonable Effectiveness of Data in Deep Learning Era](https://mlsanity.com/paper/1707.02968/) (2017)
- [PaLI: A Jointly-Scaled Multilingual Language-Image Model](https://mlsanity.com/paper/2209.06794/) (2022)
- [FILIP: Fine-grained Interactive Language-Image Pre-Training](https://mlsanity.com/paper/2111.07783/) (2021)
- [RegionCLIP: Region-based Language-Image Pretraining](https://mlsanity.com/paper/2112.09106/) (2021)
- [Extract Free Dense Labels from CLIP](https://mlsanity.com/paper/2112.01071/) (2021)

## Built on by (most cited first, 30 of 44)

- [Glyph-ByT5: A Customized Text Encoder for Accurate Visual Text Rendering](https://mlsanity.com/paper/2403.09622/) (2024)
- [Scaling Instructable Agents Across Many Simulated Worlds](https://mlsanity.com/paper/2404.10179/) (2024)
- [Softmax is not Enough (for Sharp Size Generalisation)](https://mlsanity.com/paper/2410.01104/) (2024)
- [Talking to DINO: Bridging Self-Supervised Vision Backbones with Language for Open-Vocabulary Segmentation](https://mlsanity.com/paper/2411.19331/) (2024)
- [Contrastive Localized Language-Image Pre-Training](https://mlsanity.com/paper/2410.02746/) (2024)
- [SIMA 2: A Generalist Embodied Agent for Virtual Worlds](https://mlsanity.com/paper/2512.04797/) (2025)
- [Scaling Vision Pre-Training to 4K Resolution](https://mlsanity.com/paper/2503.19903/) (2025)
- [FineLIP: Extending CLIP's Reach via Fine-Grained Alignment with Longer Text Inputs](https://mlsanity.com/paper/2504.01916/) (2025)
- [SEA: Supervised Embedding Alignment for Token-Level Visual-Textual Integration in MLLMs](https://mlsanity.com/paper/2408.11813/) (2024)
- [Preserving Multi-Modal Capabilities of Pre-trained VLMs for Improving Vision-Linguistic Compositionality](https://mlsanity.com/paper/2410.05210/) (2024)
- [Visually Descriptive Language Model for Vector Graphics Reasoning](https://mlsanity.com/paper/2404.06479/) (2024)
- [Harnessing Frozen Unimodal Encoders for Flexible Multimodal Alignment](https://mlsanity.com/paper/2409.19425/) (2024)
- [Enhancing Vision-Language Compositional Understanding with Multimodal Synthetic Data](https://mlsanity.com/paper/2503.01167/) (2025)
- [Exploring How Generative MLLMs Perceive More Than CLIP with the Same Vision Encoder](https://mlsanity.com/paper/2411.05195/) (2024)
- [β-CLIP: Text-Conditioned Contrastive Learning for Multi-Granular Vision-Language Alignment](https://mlsanity.com/paper/2512.12678/) (2025)
- [MMKU-Bench: A Multimodal Update Benchmark for Diverse Visual Knowledge](https://mlsanity.com/paper/2603.15117/) (2026)
- [Towards Real Zero-Shot Camouflaged Object Segmentation without Camouflaged Annotations](https://mlsanity.com/paper/2410.16953/) (2024)
- [GD-Retriever: Controllable Generative Text-Music Retrieval with Diffusion Models](https://mlsanity.com/paper/2506.17886/) (2025)
- [Beyond Sight: Towards Cognitive Alignment in LVLM via Enriched Visual Knowledge](https://mlsanity.com/paper/2411.16824/) (2024)
- [QuARI: Query Adaptive Retrieval Improvement](https://mlsanity.com/paper/2505.21647/) (2025)
- [Multimodal Representation Alignment for Cross-modal Information Retrieval](https://mlsanity.com/paper/2506.08774/) (2025)
- [One Patch to Caption Them All: A Unified Zero-Shot Captioning Framework](https://mlsanity.com/paper/2510.02898/) (2025)
- [Text-Conditional JEPA for Learning Semantically Rich Visual Representations](https://mlsanity.com/paper/2605.03245/) (2026)
- [HiLight: Technical Report on the Motern AI Video Language Model](https://mlsanity.com/paper/2407.07325/) (2024)
- [Grounding Descriptions in Images informs Zero-Shot Visual Recognition](https://mlsanity.com/paper/2412.04429/) (2024)
- [Multi-View Slot Attention Using Paraphrased Texts for Face Anti-Spoofing](https://mlsanity.com/paper/2509.06336/) (2025)
- [Cross-modal Full-mode Fine-grained Alignment for Text-to-Image Person Retrieval](https://mlsanity.com/paper/2509.13754/) (2025)
- [AttAnchor: Guiding Cross-Modal Token Alignment in VLMs with Attention Anchors](https://mlsanity.com/paper/2509.23109/) (2025)
- [FarSLIP: Discovering Effective CLIP Adaptation for Fine-Grained Remote Sensing Understanding](https://mlsanity.com/paper/2511.14901/) (2025)
- [ABounD: Adversarial Boundary-Driven Few-Shot Learning for Multi-Class Anomaly Detection](https://mlsanity.com/paper/2511.22436/) (2025)

## Abstract (arXiv)

We introduce SPARse Fine-grained Contrastive Alignment (SPARC), a simple method for pretraining more fine-grained multimodal representations from image-text pairs. Given that multiple image patches often correspond to single words, we propose to learn a grouping of image patches for every token in the caption. To achieve this, we use a sparse similarity metric between image patches and language tokens and compute for each token a language-grouped vision embedding as the weighted average of patches. The token and language-grouped vision embeddings are then contrasted through a fine-grained sequence-wise loss that only depends on individual samples and does not require other batch samples as negatives. This enables more detailed information to be learned in a computationally inexpensive manner. SPARC combines this fine-grained loss with a contrastive loss between global image and text embeddings to learn representations that simultaneously encode global and local information. We thoroughly evaluate our proposed method and show improved performance over competing approaches both on image-level tasks relying on coarse-grained information, e.g. classification, as well as region-level tasks relying on fine-grained information, e.g. retrieval, object detection, and segmentation. Moreover, SPARC improves model faithfulness and captioning in foundational vision-language models.