# Parrot Captions Teach CLIP to Spot Text

arXiv: https://arxiv.org/abs/2312.14232 · HTML: https://mlsanity.com/paper/2312.14232/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 41 of its 49 references that are in the map.
- First posted on arXiv 21 Dec 2023.

## Builds on (most cited first, 30 of 41)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [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)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [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)
- [GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models](https://mlsanity.com/paper/2112.10741/) (2021)
- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [Emergent Abilities of Large Language Models](https://mlsanity.com/paper/2206.07682/) (2022)
- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [Learning to Prompt for Vision-Language Models](https://mlsanity.com/paper/2109.01134/) (2021)
- [Towards VQA Models That Can Read](https://mlsanity.com/paper/1904.08920/) (2019)
- [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)
- [Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Concepts](https://mlsanity.com/paper/2102.08981/) (2021)
- [Language-driven Semantic Segmentation](https://mlsanity.com/paper/2201.03546/) (2022)
- [DataComp: In search of the next generation of multimodal datasets](https://mlsanity.com/paper/2304.14108/) (2023)
- [When and why vision-language models behave like bags-of-words, and what to do about it?](https://mlsanity.com/paper/2210.01936/) (2022)
- [LiT: Zero-Shot Transfer with Locked-image text Tuning](https://mlsanity.com/paper/2111.07991/) (2021)
- [TextCaps: a Dataset for Image Captioning with Reading Comprehension](https://mlsanity.com/paper/2003.12462/) (2020)
- [Scene Text Visual Question Answering](https://mlsanity.com/paper/1905.13648/) (2019)
- [What does CLIP know about a red circle? Visual prompt engineering for VLMs](https://mlsanity.com/paper/2304.06712/) (2023)
- [Multimodal C4: An Open, Billion-scale Corpus of Images Interleaved with Text](https://mlsanity.com/paper/2304.06939/) (2023)
- [CPT: Colorful Prompt Tuning for Pre-trained Vision-Language Models](https://mlsanity.com/paper/2109.11797/) (2021)
- [Combined Scaling for Zero-shot Transfer Learning](https://mlsanity.com/paper/2111.10050/) (2021)
- [Evaluating CLIP: Towards Characterization of Broader Capabilities and Downstream Implications](https://mlsanity.com/paper/2108.02818/) (2021)
- [ViTAEv2: Vision Transformer Advanced by Exploring Inductive Bias for Image Recognition and Beyond](https://mlsanity.com/paper/2202.10108/) (2022)
- [A Prompt Array Keeps the Bias Away: Debiasing Vision-Language Models with Adversarial Learning](https://mlsanity.com/paper/2203.11933/) (2022)

## Built on by (most cited first, 17 of 17)

- [TextAtlas5M: A Large-scale Dataset for Dense Text Image Generation](https://mlsanity.com/paper/2502.07870/) (2025)
- [Vision-centric Token Compression in Large Language Model](https://mlsanity.com/paper/2502.00791/) (2025)
- [Leveraging Visual Tokens for Extended Text Contexts in Multi-Modal Learning](https://mlsanity.com/paper/2406.02547/) (2024)
- [Skip n: A Simple Method to Reduce Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2402.01345/) (2024)
- [See the Text: From Tokenization to Visual Reading](https://mlsanity.com/paper/2510.18840/) (2025)
- [Type-R: Automatically Retouching Typos for Text-to-Image Generation](https://mlsanity.com/paper/2411.18159/) (2024)
- [Focus, Distinguish, and Prompt: Unleashing CLIP for Efficient and Flexible Scene Text Retrieval](https://mlsanity.com/paper/2408.00441/) (2024)
- [Extract Free Dense Misalignment from CLIP](https://mlsanity.com/paper/2412.18404/) (2024)
- [Improving Text Generation on Images with Synthetic Captions](https://mlsanity.com/paper/2406.00505/) (2024)
- [IntroStyle: Training-Free Introspective Style Attribution using Diffusion Features](https://mlsanity.com/paper/2412.14432/) (2024)
- [Web Artifact Attacks Disrupt Vision Language Models](https://mlsanity.com/paper/2503.13652/) (2025)
- [Exploring a Unified Vision-Centric Contrastive Alternatives on Multi-Modal Web Documents](https://mlsanity.com/paper/2510.18703/) (2025)
- [Emergence of Text Readability in Vision Language Models](https://mlsanity.com/paper/2506.19389/) (2025)
- [Turning Adversaries into Allies: Reversing Typographic Attacks for Multimodal E-Commerce Product Retrieval](https://mlsanity.com/paper/2511.05325/) (2025)
- [When Text Hijacks Vision: Benchmarking and Mitigating Text Overlay-Induced Hallucination in Vision Language Models](https://mlsanity.com/paper/2604.17375/) (2026)
- [Planning and Rendering in Concert: DeepFusion of Autoregressive Layouts and Diffusion for Visual Text Generation](https://mlsanity.com/paper/2609.22916/) (2026)
- [Vision Transformers Learn Word Meaning from Images Alone](https://mlsanity.com/paper/2610.13252/) (2026)

## Abstract (arXiv)

Despite CLIP being the foundation model in numerous vision-language applications, the CLIP suffers from a severe text spotting bias. Such bias causes CLIP models to `Parrot' the visual text embedded within images while disregarding the authentic visual semantics. We uncover that in the most popular image-text dataset LAION-2B, the captions also densely parrot (spell) the text embedded in images. Our analysis shows that around 50% of images are embedded with visual text content, and around 30% of captions words are in these embedded visual content. Based on such observation, we thoroughly inspect the different released versions of CLIP models and verify that the visual text is the dominant factor in measuring the LAION-style image-text similarity for these models. To examine whether these parrot captions shape the text spotting bias, we train a series of CLIP models with LAION subsets curated by different parrot-caption-oriented criteria. We show that training with parrot captions easily shapes such bias but harms the expected visual-language representation learning in CLIP models. This suggests that it is urgent to revisit either the design of CLIP-like models or the existing image-text dataset curation pipeline built on CLIP score filtering.