Summary of Lightweight Neural App Control, by Filippos Christianos et al.
Lightweight Neural App Control
by Filippos Christianos, Georgios Papoudakis, Thomas Coste, Jianye Hao, Jun Wang, Kun Shao
First submitted to arxiv on: 23 Oct 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper presents a novel approach for efficient mobile phone control called Lightweight Multi-modal App Control (LiMAC). LiMAC uses textual goals and past mobile observations to generate precise actions. To address computational constraints on smartphones, the authors integrate an Action Transformer (AcT) with a fine-tuned vision-language model (VLM) for real-time decision-making. The authors evaluate LiMAC on two open-source datasets, demonstrating its superior performance compared to fine-tuned VLMs and prompt engineering baselines. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary LiMAC is a new way to control your phone using text commands and past screen shots. It’s fast and efficient because it uses a special kind of AI model that can make decisions quickly. The paper shows how LiMAC works better than other methods on two big datasets. This could be important for people who want to control their phones in new ways. |
Keywords
» Artificial intelligence » Language model » Multi modal » Prompt » Transformer
