Parnian Eslamloo, Banani Roy · 2026 · arXiv (Cornell University)
DOI: 10.48550/arxiv.2610.24902 · cited by 0 works (OpenAlex) · open-access PDF
Sin embargo, los parámetros de control de errores en el contexto de Tokimat Java ofrecen datos limitados a nivel de método para estudiar la detección de excepciones en el tiempo de ejecución. Filtrando Defects4J, Bugs.jar, Bears, y BugSwarm, solo se evalúan 170 pares de control de errores reales/fixados en siete tipos de excepciones no verificadas. Introducimos RXSynth, un corpus de 4.275 pares de métodos de Java sintéticos generados a partir de ejemplares reales de errores/fix. Seleccionamos el generador utilizando alineación de dirección fina con semillas reales y validamos los pares utilizando un procedimiento de anclaje de patch en dos etapas que identifica el riesgo de excepción etiquetado y verifica si la reparación lo elimina correctamente. La
Machine translation (Meta M2M100); the English original follows.
Existing Java bug benchmarks offer limited method-level data for studying runtime-exception detection. Filtering Defects4J, Bugs.jar, Bears, and BugSwarm yields only 170 real buggy/fixed method pairs across seven unchecked exception types. We introduce RXSynth, a corpus of 4,275 synthetic Java method pairs generated from real bug/fix exemplars. We select the generator using structural alignment with real seeds and validate pairs using a two-stage patch-anchored procedure that identifies the labelled exception risk and checks whether the repair removes it. Independent JVM execution confirms patch-local validity for 84.2% of a stratified sample of 500 pairs. Provenance records enable evaluation without overlap between generation exemplars and test bugs. Structural analyses show good alignment within the represented size range, but compression of larger methods. We use RXSynth to compare isolated-method detection with pair discrimination when the repair is visible. Under provenance-controlled evaluation, supervised isolated-method ROC-AUC is about 0.55-0.64. Stronger zero-shot models reach about 69% accuracy but identify only 41.2% of repaired methods correctly. In contrast, fine-tuned encoders achieve 70.4-74.8% accuracy with both versions available. However, a simple token-growth rule achieves 84.7% expected accuracy, highlighting repair-direction cues. On 36 pairs without token growth, encoder accuracy falls to approximately chance, while GPT-5.5 and Gemini 3.1 Pro exceed 94%. RXSynth thus supports contrastive, patch-oriented, and context-aware studies; pairwise evaluations should control for provenance and directional surface cues before interpreting high accuracy as defect understanding.
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
Topics: Computer science, Java, Encoder, Pairwise comparison, Generator (circuit theory), Security token
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.24902, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange