A Primal-Dual Perspective on Program Verification Algorithms (Extended Version)
Bibliographic record
Abstract
Many algorithms in verification and automated reasoning leverage some form of duality between proofs and refutations or counterexamples. In most cases, duality is only used as an intuition that helps in understanding the algorithms and is not formalized. In other cases, duality is used explicitly, but in a specially tailored way that does not generalize to other problems. In this paper we propose a unified primal-dual framework for designing verification algorithms that leverage duality. To that end, we generalize the concept of a Lagrangian that is commonly used in linear programming and optimization to capture the domains considered in verification problems, which are usually discrete, e.g., powersets of states, predicates, ranking functions, etc. A Lagrangian then induces a primal problem and a dual problem. We devise an abstract primal-dual procedure that simultaneously searches for a primal solution and a dual solution, where the two searches guide each other. We provide sufficient conditions that ensure that the procedure makes progress under certain monotonicity assumptions on the Lagrangian. We show that many existing algorithms in program analysis, verification, and automated reasoning can be derived from our algorithmic framework with a suitable choice of Lagrangian. The Lagrangian-based formulation sheds new light on various characteristics of these algorithms, such as the ingredients they use to ensure monotonicity and guarantee progress. We further use our framework to develop a new validity checking algorithm for fixpoint logic over quantified linear arithmetic. Our prototype achieves promising results and in some cases solves instances that are not solved by state-of-the-art techniques.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.002 | 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".