# SINBAD: Saliency-informed detection of breakage caused by ad blocking

arXiv: https://arxiv.org/abs/2405.05196 · HTML: https://mlsanity.com/paper/2405.05196/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 75 references that are in the map.
- First posted on arXiv 8 May 2024.

## Builds on (most cited first, 12 of 12)

- [XGBoost: A Scalable Tree Boosting System](https://mlsanity.com/paper/1603.02754/) (2016)
- [MMDetection: Open MMLab Detection Toolbox and Benchmark](https://mlsanity.com/paper/1906.07155/) (2019)
- [Distribution-Free Predictive Inference For Regression](https://mlsanity.com/paper/1604.04173/) (2016)
- [Fingerprinting the Fingerprinters: Learning to Detect Browser Fingerprinting Behaviors](https://mlsanity.com/paper/2008.04480/) (2020)
- [WebGraph: Capturing Advertising and Tracking Information Flows for Robust Blocking](https://mlsanity.com/paper/2107.11309/) (2021)
- [Who Filters the Filters: Understanding the Growth, Usefulness and Efficiency of Crowdsourced Ad Blocking](https://mlsanity.com/paper/1810.09160/) (2018)
- [COOKIEGRAPH: Understanding and Detecting First-Party Tracking Cookies](https://mlsanity.com/paper/2208.12370/) (2022)
- [AutoFR: Automated Filter Rule Generation for Adblocking](https://mlsanity.com/paper/2202.12872/) (2022)
- [Blocked or Broken? Automatically Detecting When Privacy Interventions Break Websites](https://mlsanity.com/paper/2203.03528/) (2022)
- [Breaking Bad: Quantifying the Addiction of Web Elements to JavaScript](https://mlsanity.com/paper/2301.10597/) (2023)
- [ASTrack: Automatic Detection and Removal of Web Tracking Code with Minimal Functionality Loss](https://mlsanity.com/paper/2301.10895/) (2023)
- [Blocking JavaScript without Breaking the Web: An Empirical Investigation](https://mlsanity.com/paper/2302.01182/) (2023)

## Abstract (arXiv)

Privacy-enhancing blocking tools based on filter-list rules tend to break legitimate functionality. Filter-list maintainers could benefit from automated breakage detection tools that allow them to proactively fix problematic rules before deploying them to millions of users. We introduce SINBAD, an automated breakage detector that improves the accuracy over the state of the art by 20%, and is the first to detect dynamic breakage and breakage caused by style-oriented filter rules. The success of SINBAD is rooted in three innovations: (1) the use of user-reported breakage issues in forums that enable the creation of a high-quality dataset for training in which only breakage that users perceive as an issue is included; (2) the use of 'web saliency' to automatically identify user-relevant regions of a website on which to prioritize automated interactions aimed at triggering breakage; and (3) the analysis of webpages via subtrees which enables fine-grained identification of problematic filter rules.