Haltermann et al. present a generalization of ranged symbolic execution to arbitrary program analyses. The method, implemented on CPAchecker and tested on the SV-COMP benchmark, enhances verification efficiency by running parallel analyses. sciencedirect.com/science/articl…

Politano et al. present BEAPI, a tool generating bounded exhaustive inputs for Java classes using only the public API methods. BEAPI performs similarly to Korat but without needing a specification of the valid inputs. sciencedirect.com/science/articl…

Chen et al. introduce iGnnVD, a vulnerability detection model integrating three GNN-based detectors to improve its detection accuracy. After extensive experiments on three datasets, iGnnVD outperforms existing models in accuracy and stability. authors.elsevier.com/a/1jGBOc7X5A%7…

Zhang et al. introduce TR-Fuzz, a Transformer-based fuzzer for C compilers generating well-formed programs using position embedding and multi-head attention. TR-Fuzz outperformed existing tools and successfully discovered five bugs in GCC-9. authors.elsevier.com/a/1jHdjc7X5A%7…

Liu et al. present MATTER, a framework for comparing defect prediction models featuring a simple baseline and providing fair comparisons. Surprisingly, comparing recent models MATTER revealed that they do not outperform the baseline, as expected. authors.elsevier.com/a/1jHuT_3pyw7V…

Belenchia et al. developed libmg, a Python library for defining, visualizing, and explaining graph neural networks using a specification language μG. The use of libmg has been demonstrated by implementing a Computation Tree Logic model checker. authors.elsevier.com/a/1jHuTc7X5A%7…

Zong et al. present GraphPyRec for Python code recommendation. It converts source code into a graph and uses a child-sibling-like process and a data transfer algorithm. GraphPyRec outperforms mainstream methods on a million lines of code mined GitHub. authors.elsevier.com/a/1jKisc7X5A%7…

Lauwaerts et al. present Latch, a framework for large-scale automated testing on constrained hardware. It executes tests step-wise, skips redundant tests, and handles flaky tests. Latch is shown to be effective and reliable on an ESP32 microcontroller. authors.elsevier.com/a/1jIFic7X5A%7…

Prokić et al. introduce a prescriptive procedure for manual code smell annotation to improve the quality of datasets to train code smell detectors. It addresses issues like small dataset sizes and ambiguity and provides a tool to support annotators. authors.elsevier.com/a/1jNNYc7X5A%7…

Arcaini and Cetinkaya developed CRAG, a tool to test ADS by generating road geometries via combinatorial testing, finding scenarios where the ADS drive off-road. CRAG can be extended with different test generators, search algorithms, and test objectives. authors.elsevier.com/a/1jQSMc7X5A%7…

Bartlett et al. developed a gradient-free tool for generating adversarial examples for deep learning models using differential evolution (DE) with five algorithms. The tool has been tested on five image classification models. doi.org/10.1016/j.scic…

Cui et al. developed DPFuzz, a fuzz testing tool that prioritizes modules with higher defect proneness. DPFuzz demonstrates the practicability of using defect prediction for software quality assurance. authors.elsevier.com/a/1jTobc7X5A%7…

Banning Osei et al. propose WIDENNET to detect vulnerabilities in smart contracts (reentrancy and timestamp dependence issues) processing their bytecode as vectors. It achieved an average accuracy of 83% and a precision of 83% on real-world datasets. authors.elsevier.com/a/1jR2Xc7X5A%7…

Defourné introduces the TLA+ Proof System (TLAPS) to allow the verification of proofs via automated theorem provers. The approach is based on a first-order axiomatization with E-matching patterns. sciencedirect.com/science/articl…

Platzer evaluates three logics for hybrid dynamical systems, starting from Differential Dynamic Logic and proposing two refined variants. These logics are implemented in a theorem prover and applied to real-world cases like robot navigation and ACAS X. doi.org/10.1016/j.scic…

AlBlwi et al. unify mutant subsumption, test suite effectiveness, and program repair in software testing into one single concept: the "detector sets". It helps redefine known problems in software testing with clearer definitions. authors.elsevier.com/a/1jUr-_3pyw7V…

Pereira and Rito Silva published "Microservices Simulator," an object-oriented framework that simplifies microservice development with rapid prototyping based on the Transactional Causal Consistency model. authors.elsevier.com/a/1jYQWc7X5A%7…

Aguzzi and Viroli published MacroSwarm, an aggregate programming framework for developing swarm behaviors in Scala. With this framework, it is possible to define a wide variety of swarm behaviors, starting from simple movements to more complex ones. authors.elsevier.com/a/1jYN7c7X5A%7…

Dong et al. introduce a method for detecting infeasible paths using automatically generated program summaries. Implemented in DTSIP, it effectively identifies intraprocedural and interprocedural infeasible paths, improving testing efficiency. authors.elsevier.com/a/1jdXzc7X5A%7…

Samanipour et al. present MDAPW3 to develop Web3.0 DApps. It builds on past methods by extending beyond smart contracts, shown via a land leasing DApp. The proposed meta-models are more reusable, understandable, and offer better ontology coverage. authors.elsevier.com/a/1jeqTc7X5A%7…