# UIPress: Bringing Optical Token Compression to UI-to-Code Generation

arXiv: https://arxiv.org/abs/2604.09442 · HTML: https://mlsanity.com/paper/2604.09442/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Visual token pruning and efficiency, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 25 of its 59 references that are in the map.
- First posted on arXiv 10 Apr 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Visual token pruning and efficiency](https://mlsanity.com/mllm/visual-token-pruning-and-efficiency/)
- Efficient LLM inference and serving / [Visual token pruning in multimodal models](https://mlsanity.com/inference/visual-token-pruning-in-multimodal-models/)

## Builds on (most cited first, 25 of 25)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [MMBench: Is Your Multi-modal Model an All-around Player?](https://mlsanity.com/paper/2307.06281/) (2023)
- [MiniCPM-V: A GPT-4V Level MLLM on Your Phone](https://mlsanity.com/paper/2408.01800/) (2024)
- [GPT-4V(ision) is a Generalist Web Agent, if Grounded](https://mlsanity.com/paper/2401.01614/) (2024)
- [ClipCap: CLIP Prefix for Image Captioning](https://mlsanity.com/paper/2111.09734/) (2021)
- [Visual-RFT: Visual Reinforcement Fine-Tuning](https://mlsanity.com/paper/2503.01785/) (2025)
- [LLaVA-CoT: Let Vision Language Models Reason Step-by-Step](https://mlsanity.com/paper/2411.10440/) (2024)
- [OmniParser for Pure Vision Based GUI Agent](https://mlsanity.com/paper/2408.00203/) (2024)
- [TokenPacker: Efficient Visual Projector for Multimodal LLM](https://mlsanity.com/paper/2407.02392/) (2024)
- [GUICourse: From General Vision Language Models to Versatile GUI Agents](https://mlsanity.com/paper/2406.11317/) (2024)
- [GUI-G2: Gaussian Reward Modeling for GUI Grounding](https://mlsanity.com/paper/2507.15846/) (2025)
- [Foundations and Recent Trends in Multimodal Mobile Agents: A Survey](https://mlsanity.com/paper/2411.02006/) (2024)
- [Cooperation on the Fly: Exploring Language Agents for Ad Hoc Teamwork in the Avalon Game](https://mlsanity.com/paper/2312.17515/) (2023)
- [GUI-Reflection: Empowering Multimodal GUI Models with Self-Reflection Behavior](https://mlsanity.com/paper/2506.08012/) (2025)
- [Monte Carlo Planning with Large Language Model for Text-Based Game Agents](https://mlsanity.com/paper/2504.16855/) (2025)
- [Infinity Parser: Layout Aware Reinforcement Learning for Scanned Document Parsing](https://mlsanity.com/paper/2510.15349/) (2025)
- [E-ANT: A Large-Scale Dataset for Efficient Automatic GUI NavigaTion](https://mlsanity.com/paper/2406.14250/) (2024)
- [Smoothing Grounding and Reasoning for MLLM-Powered GUI Agents with Query-Oriented Pivot Tasks](https://mlsanity.com/paper/2503.00401/) (2025)
- [MMCLIP: Cross-modal Attention Masked Modelling for Medical Language-Image Pre-Training](https://mlsanity.com/paper/2407.19546/) (2024)
- [Automotive-ENV: Benchmarking Multimodal Agents in Vehicle Interface Systems](https://mlsanity.com/paper/2509.21143/) (2025)

## Built on by (most cited first, 1 of 1)

- [Explicit Representation Alignment for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2606.09148/) (2026)

## Abstract (arXiv)

UI-to-Code generation requires vision-language models (VLMs) to produce thousands of tokens of structured HTML/CSS from a single screenshot, making visual token efficiency critical. Existing compression methods either select tokens at inference time using task-agnostic heuristics, or zero out low-attention features without actually shortening the sequence -- neither truly reduces prefill latency or adapts to the non-uniform information density of UI screenshots. Meanwhile, optical (encoder-side learned) compression has shown strong results for document OCR, yet no prior work has adapted this paradigm to UI-to-Code generation. We propose UIPress, a lightweight learned compression module inserted between the frozen ViT encoder and the LLM decoder of Qwen3-VL-8B. UIPress combines depthwise-separable convolutions, element-guided spatial reweighting, and Transformer refinement to compress ~6,700 visual tokens to a fixed budget of 256. Together with Low-Rank Adaptation (LoRA) on the decoder to bridge the representation gap, the entire system adds only ~21.7M trainable parameters (0.26% of the 8B base model). Under a fair comparison on the same base model against four baselines on Design2Code, UIPress at 256 tokens achieves a CLIP score of 0.8127, outperforming the uncompressed baseline by +7.5% and the strongest inference-time method by +4.6%, while delivering 9.1× time-to-first-token speedup. To the best of our knowledge, UIPress is the first encoder-side learned compression method for the UI-to-Code task.