Bibliographic record
Abstract
Runtime monitoring is an increasingly popular method to ensure the safe execution of untrusted codes. Monitors observe and transform the execution of these codes, responding when needed to correct or prevent a violation of a user-defined security policy. Prior research has shown that the set of properties monitors can enforce correlates with the latitude they are given to transform and alter the target execution. But for enforcement to be meaningful this capacity must be constrained, otherwise the monitor can enforce any property, but not necessarily in a manner that is useful or desirable. However, such constraints have not been significantly addressed in prior work. In this article, we develop a new paradigm of security policy enforcement in which the behavior of the enforcement mechanism is restricted to ensure that valid aspects present in the execution are preserved notwithstanding any transformation it may perform. These restrictions capture the desired behavior of valid executions of the program, and are stated by way of a preorder over sequences. The resulting model is closer than previous ones to what would be expected of a real-life monitor, from which we demand a minimal footprint on both valid and invalid executions. We illustrate this framework with examples of real-life security properties. Since several different enforcement alternatives of the same property are made possible by the flexibility of this type of enforcement, our study also provides metrics that allow the user to compare monitors objectively and choose the best enforcement paradigm for a given situation.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.007 | 0.037 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.019 | 0.005 |
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 source (direct Gemma or distilled Codex), 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".