# Linguistic Binding in Diffusion Models: Enhancing Attribute Correspondence through Attention Map Alignment

arXiv: https://arxiv.org/abs/2306.08877 · HTML: https://mlsanity.com/paper/2306.08877/

## Key facts

- Cited by 171 papers in the map (data through 14 Oct 2026).
- Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models.
- Builds on 21 of its 24 references that are in the map.
- First posted on arXiv 15 Jun 2023.

## Placed in

- Diffusion models / [Image generation, text-to-image and editing](https://mlsanity.com/diffusion/image-generation-text-to-image-and-editing/)

## Builds on (most cited first, 21 of 21)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Prompt-to-Prompt Image Editing with Cross Attention Control](https://mlsanity.com/paper/2208.01626/) (2022)
- [An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion](https://mlsanity.com/paper/2208.01618/) (2022)
- [CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning](https://mlsanity.com/paper/1612.06890/) (2016)
- [eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoisers](https://mlsanity.com/paper/2211.01324/) (2022)
- [Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2301.13826/) (2023)
- [Compositional Visual Generation with Composable Diffusion Models](https://mlsanity.com/paper/2206.01714/) (2022)
- [When and why vision-language models behave like bags-of-words, and what to do about it?](https://mlsanity.com/paper/2210.01936/) (2022)
- [Training-Free Structured Diffusion Guidance for Compositional Text-to-Image Synthesis](https://mlsanity.com/paper/2212.05032/) (2022)
- [Key-Locked Rank One Editing for Text-to-Image Personalization](https://mlsanity.com/paper/2305.01644/) (2023)
- [Open World Compositional Zero-Shot Learning](https://mlsanity.com/paper/2101.12609/) (2021)
- [A causal view of compositional zero-shot recognition](https://mlsanity.com/paper/2006.14610/) (2020)
- [Testing Relational Understanding in Text-Guided Image Generation](https://mlsanity.com/paper/2208.00005/) (2022)
- [DALLE-2 is Seeing Double: Flaws in Word-to-Concept Mapping in Text2Image Models](https://mlsanity.com/paper/2210.10606/) (2022)
- [Harnessing the Spatial-Temporal Attention of Diffusion Models for High-Fidelity Text-to-Image Synthesis](https://mlsanity.com/paper/2304.03869/) (2023)
- [DALL-E 2 Fails to Reliably Capture Common Syntactic Processes](https://mlsanity.com/paper/2210.12889/) (2022)
- [Learning to generalize to new compositions in image understanding](https://mlsanity.com/paper/1608.07639/) (2016)

## Built on by (most cited first, 30 of 171)

- [ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignment](https://mlsanity.com/paper/2403.05135/) (2024)
- [T2I-CompBench++: An Enhanced and Comprehensive Benchmark for Compositional Text-to-image Generation](https://mlsanity.com/paper/2307.06350/) (2023)
- [Mastering Text-to-Image Diffusion: Recaptioning, Planning, and Generating with Multimodal LLMs](https://mlsanity.com/paper/2401.11708/) (2024)
- [Subject-Diffusion:Open Domain Personalized Text-to-Image Generation without Test-time Fine-tuning](https://mlsanity.com/paper/2307.11410/) (2023)
- [T2V-CompBench: A Comprehensive Benchmark for Compositional Text-to-video Generation](https://mlsanity.com/paper/2407.14505/) (2024)
- [Cross-Image Attention for Zero-Shot Appearance Transfer](https://mlsanity.com/paper/2311.03335/) (2023)
- [Controllable Generation with Text-to-Image Diffusion Models: A Survey](https://mlsanity.com/paper/2403.04279/) (2024)
- [Diffusion Model is Secretly a Training-free Open Vocabulary Semantic Segmenter](https://mlsanity.com/paper/2309.02773/) (2023)
- [Be Yourself: Bounded Attention for Multi-Subject Text-to-Image Generation](https://mlsanity.com/paper/2403.16990/) (2024)
- [CoMat: Aligning Text-to-Image Diffusion Model with Image-to-Text Concept Matching](https://mlsanity.com/paper/2404.03653/) (2024)
- [DreamSync: Aligning Text-to-Image Generation with Image Understanding Feedback](https://mlsanity.com/paper/2311.17946/) (2023)
- [Make It Count: Text-to-Image Generation with an Accurate Number of Objects](https://mlsanity.com/paper/2406.10210/) (2024)
- [TC-Bench: Benchmarking Temporal Compositionality in Text-to-Video and Image-to-Video Generation](https://mlsanity.com/paper/2406.08656/) (2024)
- [A Picture is Worth a Thousand Words: Principled Recaptioning Improves Image Generation](https://mlsanity.com/paper/2310.16656/) (2023)
- [Continuous, Subject-Specific Attribute Control in T2I Models by Identifying Semantic Directions](https://mlsanity.com/paper/2403.17064/) (2024)
- [TokenCompose: Text-to-Image Diffusion with Token-level Supervision](https://mlsanity.com/paper/2312.03626/) (2023)
- [Interpretable Diffusion via Information Decomposition](https://mlsanity.com/paper/2310.07972/) (2023)
- [PALP: Prompt Aligned Personalization of Text-to-Image Models](https://mlsanity.com/paper/2401.06105/) (2024)
- [Emergence of Hidden Capabilities: Exploring Learning Dynamics in Concept Space](https://mlsanity.com/paper/2406.19370/) (2024)
- [Alignment of Diffusion Models: Fundamentals, Challenges, and Future](https://mlsanity.com/paper/2409.07253/) (2024)
- [Token Merging for Training-Free Semantic Binding in Text-to-Image Synthesis](https://mlsanity.com/paper/2411.07132/) (2024)
- [A Survey on Mechanistic Interpretability for Multi-Modal Foundation Models](https://mlsanity.com/paper/2502.17516/) (2025)
- [T2I-Copilot: A Training-Free Multi-Agent Text-to-Image System for Enhanced Prompt Interpretation and Interactive Generation](https://mlsanity.com/paper/2507.20536/) (2025)
- [Prioritized Semantic Learning for Zero-shot Instance Navigation](https://mlsanity.com/paper/2403.11650/) (2024)
- [CosmicMan: A Text-to-Image Foundation Model for Humans](https://mlsanity.com/paper/2404.01294/) (2024)
- [Rare-to-Frequent: Unlocking Compositional Generation Power of Diffusion Models on Rare Concepts with LLM Guidance](https://mlsanity.com/paper/2410.22376/) (2024)
- [ColorPeel: Color Prompt Learning with Diffusion Models via Color and Shape Disentanglement](https://mlsanity.com/paper/2407.07197/) (2024)
- [Enhancing Semantic Fidelity in Text-to-Image Synthesis: Attention Regulation in Diffusion Models](https://mlsanity.com/paper/2403.06381/) (2024)
- [MC2: Multi-concept Guidance for Customized Multi-concept Generation](https://mlsanity.com/paper/2404.05268/) (2024)
- [A Survey of Diffusion Based Image Generation Models: Issues and Their Solutions](https://mlsanity.com/paper/2308.13142/) (2023)

## Abstract (arXiv)

Text-conditioned image generation models often generate incorrect associations between entities and their visual attributes. This reflects an impaired mapping between linguistic binding of entities and modifiers in the prompt and visual binding of the corresponding elements in the generated image. As one notable example, a query like "a pink sunflower and a yellow flamingo" may incorrectly produce an image of a yellow sunflower and a pink flamingo. To remedy this issue, we propose SynGen, an approach which first syntactically analyses the prompt to identify entities and their modifiers, and then uses a novel loss function that encourages the cross-attention maps to agree with the linguistic binding reflected by the syntax. Specifically, we encourage large overlap between attention maps of entities and their modifiers, and small overlap with other entities and modifier words. The loss is optimized during inference, without retraining or fine-tuning the model. Human evaluation on three datasets, including one new and challenging set, demonstrate significant improvements of SynGen compared with current state of the art methods. This work highlights how making use of sentence structure during inference can efficiently and substantially improve the faithfulness of text-to-image generation.