# Giraffe: A Mapping Architecture from Hidden Text Representations to Visual Embeddings for Efficient Graphic Design

arXiv: https://arxiv.org/abs/2608.23970 · HTML: https://mlsanity.com/paper/2608.23970/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Visual token pruning and efficiency, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 20 of its 25 references that are in the map.
- First posted on arXiv 25 Aug 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Visual token pruning and efficiency](https://mlsanity.com/mllm/visual-token-pruning-and-efficiency/)

## Builds on (most cited first, 20 of 20)

- [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)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [DreamFusion: Text-to-3D using 2D Diffusion](https://mlsanity.com/paper/2209.14988/) (2022)
- [Gemma 3 Technical Report](https://mlsanity.com/paper/2503.19786/) (2025)
- [Chameleon: Mixed-Modal Early-Fusion Foundation Models](https://mlsanity.com/paper/2405.09818/) (2024)
- [OFA: Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Framework](https://mlsanity.com/paper/2202.03052/) (2022)
- [Multimodal Few-Shot Learning with Frozen Language Models](https://mlsanity.com/paper/2106.13884/) (2021)
- [Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model](https://mlsanity.com/paper/2408.11039/) (2024)
- [DreamLLM: Synergistic Multimodal Comprehension and Creation](https://mlsanity.com/paper/2309.11499/) (2023)
- [Generating Images with Multimodal Language Models](https://mlsanity.com/paper/2305.17216/) (2023)
- [Grounding Language Models to Images for Multimodal Inputs and Outputs](https://mlsanity.com/paper/2301.13823/) (2023)
- [UniTAB: Unifying Text and Box Outputs for Grounded Vision-Language Modeling](https://mlsanity.com/paper/2111.12085/) (2021)
- [MAGMA -- Multimodal Augmentation of Generative Models through Adapter-based Finetuning](https://mlsanity.com/paper/2112.05253/) (2021)
- [Vitron: A Unified Pixel-level Vision LLM for Understanding, Generating, Segmenting, Editing](https://mlsanity.com/paper/2412.19806/) (2024)
- [Towards Perceiving Small Visual Details in Zero-shot Visual Question Answering with Multimodal LLMs](https://mlsanity.com/paper/2310.16033/) (2023)

## Abstract (arXiv)

Multimodal large language models (MLLMs) have made significant progress in understanding and interpreting mul- timedia content. However, their ability to generate me- dia remains limited. Recent approaches have attempted to bridge this gap by translating the hidden representations of token sequences into the embedding space of visual models or directly into raw image data. However, these methods often represent each image using multiple specialised to- kens which significantly increases the input length. This be- comes a major limitation for tasks such as graphic design generation where the output typically involves a seamless blend of thousands of tokens across text, multiple images, and layout information. To address this challenge, a novel architecture is proposed that maps hidden token represen- tations to the embedding space of visual models, such as CLIP ViT-L/14, using a single [IMG] token per image. The architecture employs two shallow MLP blocks, each with a separate compression module followed by a shared expan- sion module, trained with six distinct loss functions. One block aids the other during training and is omitted during inference, resulting in a lightweight solution. Strong perfor- mance is demonstrated in both image-to-design and text-to- design generation tasks.