LLM-synthesized GPU query kernels outperform engineered engines by 7.4x, but a portable SYCL engine with lifted optimizations closes the gap to 1.27x, suggesting engineering remains preferable on GPUs.
Paul Boniol, Qinghua Liu, Mingyi Huang, Themis Palpanas, and John Paparrizos
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4representative citing papers
A 0.3M-parameter patch-embedding method with a memory bank of normal patches beats Transformers and foundation models on TSB-AD univariate and multivariate anomaly detection.
CoAD unifies outlier exposure classification and masked autoencoder reconstruction in a cooperative loop to detect subtle and prolonged time series anomalies.
PaAno+ extends the original PaAno with multiscale feature extraction, cross-variable fusion attention, and a temporal patch sorting pretext task to report state-of-the-art results on the TSB-AD benchmark for univariate and multivariate anomaly detection.
citing papers explorer
-
From Custom-Fit to Portable: Bridging the Gap Between Synthesized and Engineered GPU Query Execution
LLM-synthesized GPU query kernels outperform engineered engines by 7.4x, but a portable SYCL engine with lifted optimizations closes the gap to 1.27x, suggesting engineering remains preferable on GPUs.
-
PaAno: Patch-Based Representation Learning for Time-Series Anomaly Detection
A 0.3M-parameter patch-embedding method with a memory bank of normal patches beats Transformers and foundation models on TSB-AD univariate and multivariate anomaly detection.
-
Bridging Classification and Reconstruction: Cooperative Time Series Anomaly Detection
CoAD unifies outlier exposure classification and masked autoencoder reconstruction in a cooperative loop to detect subtle and prolonged time series anomalies.
-
PaAno+: Multiscale Encoding and Cross-Variable Attention for Time Series Anomaly Detection
PaAno+ extends the original PaAno with multiscale feature extraction, cross-variable fusion attention, and a temporal patch sorting pretext task to report state-of-the-art results on the TSB-AD benchmark for univariate and multivariate anomaly detection.