# UI-JEPA: Towards Active Perception of User Intent through Onscreen User Activity

arXiv: https://arxiv.org/abs/2409.04081 · HTML: https://mlsanity.com/paper/2409.04081/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 38 references that are in the map.
- First posted on arXiv 6 Sep 2024.

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

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training](https://mlsanity.com/paper/2203.12602/) (2022)
- [Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture](https://mlsanity.com/paper/2301.08243/) (2023)
- [AutoDroid: LLM-powered Task Automation in Android](https://mlsanity.com/paper/2308.15272/) (2023)
- [Ferret-UI: Grounded Mobile UI Understanding with Multimodal LLMs](https://mlsanity.com/paper/2404.05719/) (2024)
- [Screen2Words: Automatic Mobile UI Summarization with Multimodal Learning](https://mlsanity.com/paper/2108.03353/) (2021)
- [Meta-Prompting: Enhancing Language Models with Task-Agnostic Scaffolding](https://mlsanity.com/paper/2401.12954/) (2024)
- [ScreenAI: A Vision-Language Model for UI and Infographics Understanding](https://mlsanity.com/paper/2402.04615/) (2024)
- [UIBert: Learning Generic Multimodal Representations for UI Understanding](https://mlsanity.com/paper/2107.13731/) (2021)
- [AutoAct: Automatic Agent Learning from Scratch for QA via Self-Planning](https://mlsanity.com/paper/2401.05268/) (2024)
- [Spotlight: Mobile UI Understanding using Vision-Language Models with a Focus](https://mlsanity.com/paper/2209.14927/) (2022)
- [Screen Recognition: Creating Accessibility Metadata for Mobile Applications from Pixels](https://mlsanity.com/paper/2101.04893/) (2021)
- [ActionBert: Leveraging User Actions for Semantic Understanding of User Interfaces](https://mlsanity.com/paper/2012.12350/) (2020)
- [Screen2Vec: Semantic Embedding of GUI Screens and GUI Components](https://mlsanity.com/paper/2101.11103/) (2021)
- [AXNav: Replaying Accessibility Tests from Natural Language](https://mlsanity.com/paper/2310.02424/) (2023)
- [Automatic Macro Mining from Interaction Traces at Scale](https://mlsanity.com/paper/2310.07023/) (2023)
- [Never-ending Learning of User Interfaces](https://mlsanity.com/paper/2308.08726/) (2023)

## Built on by (most cited first, 4 of 4)

- [Safeguarding Vision-Language Models: Mitigating Vulnerabilities to Gaussian Noise in Perturbation-based Attacks](https://mlsanity.com/paper/2504.01308/) (2025)
- [SV3.3B: A Sports Video Understanding Model for Action Recognition](https://mlsanity.com/paper/2507.17844/) (2025)
- [Bridging Values and Behavior: A Hierarchical Framework for Proactive Embodied Agents](https://mlsanity.com/paper/2604.27699/) (2026)
- [Distributed JEPA: A Self-Supervised Framework for Energy Forecasting](https://mlsanity.com/paper/2609.17029/) (2026)

## Abstract (arXiv)

Generating user intent from a sequence of user interface (UI) actions is a core challenge in comprehensive UI understanding. Recent advancements in multimodal large language models (MLLMs) have led to substantial progress in this area, but their demands for extensive model parameters, computing power, and high latency makes them impractical for scenarios requiring lightweight, on-device solutions with low latency or heightened privacy. Additionally, the lack of high-quality datasets has hindered the development of such lightweight models. To address these challenges, we propose UI-JEPA, a novel framework that employs masking strategies to learn abstract UI embeddings from unlabeled data through self-supervised learning, combined with an LLM decoder fine-tuned for user intent prediction. We also introduce two new UI-grounded multimodal datasets, "Intent in the Wild" (IIW) and "Intent in the Tame" (IIT), designed for few-shot and zero-shot UI understanding tasks. IIW consists of 1.7K videos across 219 intent categories, while IIT contains 914 videos across 10 categories. We establish the first baselines for these datasets, showing that representations learned using a JEPA-style objective, combined with an LLM decoder, can achieve user intent predictions that match the performance of state-of-the-art large MLLMs, but with significantly reduced annotation and deployment resources. Measured by intent similarity scores, UI-JEPA outperforms GPT-4 Turbo and Claude 3.5 Sonnet by 10.0% and 7.2% respectively, averaged across two datasets. Notably, UI-JEPA accomplishes the performance with a 50.5x reduction in computational cost and a 6.6x improvement in latency in the IIW dataset. These results underscore the effectiveness of UI-JEPA, highlighting its potential for lightweight, high-performance UI understanding.