Problem
Research questions and friction points this paper is trying to address.
mobile ads detection
dynamic content
UI rendering
ad widget identification
runtime ad triggering
From UI to Code: Mobile Ads Detection via LLM-Unified Static-Dynamic Analysis
🤖 AI Summary
Mobile advertisements are challenging to accurately locate and trigger at runtime due to their dynamic loading behavior and diverse user interface (UI) structures. This work proposes ADWISE, a novel framework that uniquely integrates static program analysis with large language model (LLM)-driven dynamic UI exploration. ADWISE employs a tripartite guidance mechanism—comprising window transition graph (WTG)-based global navigation priors, semantic functionality reasoning, and retrieval-augmented layout matching—to efficiently identify and interact with ad controls. Evaluated on a benchmark of 100 real-world applications, ADWISE improves ad detection accuracy by 25.60% over the state-of-the-art method and uncovers an additional 34.34% of non-compliant ads, spanning six distinct categories of regulatory violations.
📝 Abstract
Mobile advertisements (ads) are essential to the app economy, yet detecting them is challenging because ad content is dynamically fetched from remote servers and rendered through diverse user interfaces (UIs), making ads difficult to locate and trigger at runtime. To address this challenge, we present ADWISE, a novel framework that formulates mobile ads detection as LLM-guided, ad-oriented UI exploration. ADWISE first performs static program analysis to identify UI widgets used to place ads, which we call ad widgets. It then uses a grounded LLM reasoning loop to navigate toward and trigger these widgets under three complementary domain guidance signals: (1) WTG-based guidance, which provides global transition priors from a statically constructed window transition graph (WTG); (2) semantic guidance, which reasons over app functionality to prioritize user-likely interaction paths; and (3) structural guidance, which applies retrieval-augmented generation to match the current UI against recurring ad-heavy layouts from a knowledge base. By combining static program analysis with LLM-based reasoning over UI structure, app semantics, and retrieved analogies, ADWISE enables more effective ads detection in complex mobile UIs. Experiments on 100 benchmark apps show that ADWISE outperforms state-of-the-art baselines by 25.60% in ad widget detection. In addition, ADWISE uncovers 34.34% more ad regulation violations across six categories, directly benefiting downstream ad regulation.
Innovation
Methods, ideas, or system contributions that make the work stand out.
LLM-guided UI exploration
static-dynamic analysis
ad widget detection
retrieval-augmented generation
window transition graph
🔎 Similar Papers
No similar papers found.
