MANA: Towards Efficient Mobile Ad Detection via Multimodal Agentic UI Navigation

• URL: http://arxiv.org/abs/2603.20351v1
• Date: Fri, 20 Mar 2026 12:19:41 GMT
• Title: MANA: Towards Efficient Mobile Ad Detection via Multimodal Agentic UI Navigation
• Abstract summary: We present MANA, the first agentic multimodal reasoning framework for mobile ad detection.<n>We implement and evaluate MANA on commercial smartphones over 200 apps, achieving state-of-the-art accuracy and efficiency.
• Score: 28.21505173665395
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Mobile advertising dominates app monetization but introduces risks ranging from intrusive user experience to malware delivery. Existing detection methods rely either on static analysis, which misses runtime behaviors, or on heuristic UI exploration, which struggles with sparse and obfuscated ads. In this paper, we present MANA, the first agentic multimodal reasoning framework for mobile ad detection. MANA integrates static, visual, temporal, and experiential signals into a reasoning-guided navigation strategy that determines not only how to traverse interfaces but also where to focus, enabling efficient and robust exploration. We implement and evaluate MANA on commercial smartphones over 200 apps, achieving state-of-the-art accuracy and efficiency. Compared to baselines, it improves detection accuracy by 30.5%-56.3% and reduces exploration steps by 29.7%-63.3%. Case studies further demonstrate its ability to uncover obfuscated and malicious ads, underscoring its practicality for mobile ad auditing and its potential for broader runtime UI analysis (e.g., permission abuse). Code and dataset are available at https://github.com/MANA-2026/MANA.

Related papers

• When Ad Networks Misbehave: Understanding Risks of Semi-Drive-By Splash Ads [13.802323882269995]
  We investigate the mobile splash ads ecosystem, where monetization relies on interaction signals that are difficult to verify end-to-end.<n>We uncover a previously less-known form of fraud at the ad-network layer in which splash ads are triggered not by intentional user actions but by incidental or indirect interactions.<n>By translating non-ad interactions into billable engagement events, ad networks can inflate performance metrics, overcharge advertisers, and erode user trust.
  arXiv  Detail & Related papers  (2026-09-09T00:54:43Z)
• SPIEval: Evaluating Large Language Models as Mobile Assistants over Scattered Personal Information [61.41881137371855]
  We introduce SPIEval, a benchmark for large language models (LLMs)<n> SPIEval comprises 250 tasks spanning 4,335 personal records distributed across 10 apps and supports multi-turn interaction through 21 tools.<n>The best-performing model, GPT-5.5 (xhigh), achieves only 57.3% accuracy, while the weakest achieves just 16.4%.
  arXiv  Detail & Related papers  (2026-08-11T09:14:53Z)
• Expecting (Targeted Ads)? Network Analysis of User Health Data Leakage in Fertility Tracking Apps [9.426024604008406]
  We conduct a network-based measurement study of a corpus of 20 Android fertility tracking apps from the Google Play Store.<n>We identify explicit leakage of user health data as well implicit leakage through highly targeted contextual advertising URL's.<n>These findings provide technical grounding for widespread user concerns, but also underscore the importance of consumer choice in the privacy implications of app-based fertility tracking.
  arXiv  Detail & Related papers  (2026-06-24T18:17:04Z)
• From UI to Code: Mobile Ads Detection via LLM-Unified Static-Dynamic Analysis [21.13861921564872]
  ADWISE is a framework that formulates mobile ads detection as LLM-guided, ad-oriented UI exploration.<n>ADWISE first performs static program analysis to identify UI widgets used to place ads, which we call ad widgets.<n>ADWISE outperforms state-of-the-art baselines by 25.60% in ad widget detection.
  arXiv  Detail & Related papers  (2026-04-04T03:04:03Z)
• CovAgent: Overcoming the 30% Curse of Mobile Application Coverage with Agentic AI and Dynamic Instrumentation [10.80010959571188]
  CovAgent is a novel agentic AI-powered approach to enhance Android app UI testing.<n>Our framework achieves a significant improvement in test coverage over the state-of-the-art, LLMDroid.<n>CovAgent also outperforms all the baselines in other metrics such as class, method, and line coverage.
  arXiv  Detail & Related papers  (2026-01-29T04:21:11Z)
• Mobile-Agent-V: A Video-Guided Approach for Effortless and Efficient Operational Knowledge Injection in Mobile Automation [53.54951412651823]
  Mobile-Agent-V is an innovative framework that utilizes video as a guiding tool to effortlessly and efficiently inject operational knowledge into mobile automation processes.<n>By deriving knowledge directly from video content, Mobile-Agent-V eliminates manual intervention, significantly reducing the effort and time required for knowledge acquisition.<n>Our experimental findings demonstrate that Mobile-Agent-V enhances performance by 36% compared to existing methods.
  arXiv  Detail & Related papers  (2025-05-20T03:48:19Z)
• Mobile-Agent-V: A Video-Guided Approach for Effortless and Efficient Operational Knowledge Injection in Mobile Automation [53.54951412651823]
  Mobile-Agent-V is an innovative framework that utilizes video as a guiding tool to effortlessly and efficiently inject operational knowledge into mobile automation processes.<n>By deriving knowledge directly from video content, Mobile-Agent-V eliminates manual intervention, significantly reducing the effort and time required for knowledge acquisition.<n>Our experimental findings demonstrate that Mobile-Agent-V enhances performance by 36% compared to existing methods.
  arXiv  Detail & Related papers  (2025-02-24T12:51:23Z)
• Auto-Intent: Automated Intent Discovery and Self-Exploration for Large Language Model Web Agents [68.22496852535937]
  We introduce Auto-Intent, a method to adapt a pre-trained large language model (LLM) as an agent for a target domain without direct fine-tuning. Our approach first discovers the underlying intents from target domain demonstrations unsupervisedly. We train our intent predictor to predict the next intent given the agent's past observations and actions.
  arXiv  Detail & Related papers  (2024-10-29T21:37:04Z)
• AppPoet: Large Language Model based Android malware detection via multi-view prompt engineering [1.3197408989895103]
  AppPoet is a multi-view system for Android malware detection. Our method achieves a detection accuracy of 97.15% and an F1 score of 97.21%, which is superior to the baseline methods.
  arXiv  Detail & Related papers  (2024-04-29T15:52:45Z)
• Mask Off: Analytic-based Malware Detection By Transfer Learning and Model Personalization [7.717537870226507]
  Analytic-based deep neural network, Android Malware detection (ADAM) This paper presents an Analytic-based deep neural network, Android Malware detection (ADAM)
  arXiv  Detail & Related papers  (2022-11-20T01:52:15Z)
• Domain and Modality Gaps for LiDAR-based Person Detection on Mobile Robots [91.01747068273666]
  This paper studies existing LiDAR-based person detectors with a particular focus on mobile robot scenarios. Experiments revolve around the domain gap between driving and mobile robot scenarios, as well as the modality gap between 3D and 2D LiDAR sensors. Results provide practical insights into LiDAR-based person detection and facilitate informed decisions for relevant mobile robot designs and applications.
  arXiv  Detail & Related papers  (2021-06-21T16:35:49Z)
• Active Visual Information Gathering for Vision-Language Navigation [115.40768457718325]
  Vision-language navigation (VLN) is the task of entailing an agent to carry out navigational instructions inside photo-realistic environments. One of the key challenges in VLN is how to conduct a robust navigation by mitigating the uncertainty caused by ambiguous instructions and insufficient observation of the environment. This work draws inspiration from human navigation behavior and endows an agent with an active information gathering ability for a more intelligent VLN policy.
  arXiv  Detail & Related papers  (2020-07-15T23:54:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.