Brief

Researchers have developed a novel "Cluster-then-Summarize" pipeline to automate the generation of test specifications for large-scale automotive software requirements. This method embeds requirements, clusters them using UMAP and HDBSCAN, and then summarizes each cluster to preserve critical information. The pipeline generates tests at both individual requirement and cluster-level integration points, enhancing coverage and efficiency compared to standard LLM approaches. AI

Researchers have developed ClustRecNet, a novel deep learning framework designed to automatically recommend effective clustering algorithms for datasets. This end-to-end system learns directly from raw tabular data, bypassing the need for manual feature engineering. ClustRecNet was trained on a large repository of synthetic datasets and demonstrated superior performance over traditional validity indices and existing AutoML approaches on both synthetic and real-world benchmarks. AI