Static Program Slicing Using Language Models With Dataflow-Aware Pretraining and Constrained Decoding
2026-05-01T08:51:54Z•afcd8c601cbbc01869953d629d7dd8e69ef9194d60387f7b8487b2be75b6f984
CI-Repair-BenchCI-reproducibilityEvoSuiteLLM-test-generationLLM-variabilityautomated-patch-validationbuild-reproductionconstrained-decodingdataflow-aware-pretrainingfeedback-driven-fuzzingmachine-generated-code-detectionmodular-testingresearch-software-orchestrationsecurity-testingsoftware-supply-chainstatic-program-slicing
What happened
Collection of recent CS-Software Engineering papers with direct relevance to software security and developer toolchains. Key contributions: Sliceformer — dataflow-aware pretraining and constrained decoding for more precise static program slicing (Java/Python), improving slice accuracy and reducing hallucination; a SemEval system for robust detection/attribution of machine-generated code and analysis of severe class-imbalance effects; a systematic survey of adaptive and AI-augmented security testing that identifies a structural-adaptive fragmentation and outlines a unified agenda for semmantics
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- arxiv_cs_se
- Record identifier
- afcd8c601cbbc01869953d629d7dd8e69ef9194d60387f7b8487b2be75b6f984
- Enrichment time
- 2026-05-01T08:51:54Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.