Bibliographic record
Abstract
Android devices remain an attractive mobile malware target in recent years. Android applications (or simply apps) in the device are vulnerable to different attacks which can tamper with the execution of an app to change app behavior so that it performs harm to users or can debug an app to steal private data (source code, user data and behavior). Android app protection is necessary to defend app behavior integrity and protect app privacy. \n \nThe app cache, where the app actually runs, is vulnerable to being tampered with. Cache tampering allows for the same behavioral changes as piggybacking. Piggybacking an app is to repackage an legitimate app with extra code that can perform malicious acts after installation, such as stealing user sensitive data or displaying unsolicited advertisements. The cache loading process of Android Runtime (ART) can be exploited by cache tampering attacks without rebooting the device. \nSecurity-Enhanced Linux (SELinux) full enforcement has been deployed in the Android platform since Android 5, which enhances the security of Android platform and decreases the security concerns apps should take care of at the same time. Therefore, apps are vulnerable to being debugged in an insecure Android environment such as an emulator or a device with a rooted Android ROM. \n \nWe present a comprehensive app protection approach using anti-tampering and anti-debugging techniques. We implement separate solutions in terms of two protections against tampering and debugging. We maintain the integrity of app cache \nand implement a lightweight cache protection solution for anti-tampering. We collect debugging points of ART and protect them at runtime from being tampered with. Our solution can be deployed easily across different Android ART-based platforms with little effort. App developers are able to use our techniques to protect their apps.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.006 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".