Tegawendé F. Bissyandé

Neural Machine Translation for Moor´e, a Low-Resource Language

Secrets Unlocked: Evaluating LLMs for Secrets Detection in Android Apps

Mobile apps frequently embed sensitive secrets, such as API keys, access tokens, client secrets, and private keys that support internal functionality or enable integration with external systems and third-party services. Developers frequently embed these secrets into Android apps, which allows attackers to extract them...

Test input prioritization for image segmentation: an empirical study

This study evaluates 26 methods across nine models and three datasets, analyzing factors such as IoU thresholds, embedding models, feature fusion techniques, and key parameters, and finds that complexity-based and learning-based methods consistently perform best across all settings.

Semantic Drift in Bug Resolution: How Behavioral Signals Propagate from Reports to Tests and Patches

This analysis covers 2,857 report-test-patch triplets from Defects4J and SWT-Bench using two widely adopted instruction-tuned LLMs from distinct model families and finds that both models exhibit systematic optimism relative to humans and only modest rank agreement, motivating bias-aware evaluation.

L\"etzCross: A Cross-Lingual Page-Level Benchmark for Multimodal Retrieval over Luxembourgish Documents

This work introduces L\"etzCross, a benchmark for cross-lingual page-level retrieval over Luxembourgish PDF documents, with document pages indexed as images and queries provided in English, French, German, and Luxembourgish, and examines single-language and multilingual fine-tuning.