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.
Title resolution pending
5 Pith papers cite this work, alongside 4 external citations. Polarity classification is still indexing.
5
Pith papers citing it
4
external citations · external index
citation-role summary
dataset 1
citation-polarity summary
years
2026 5roles
dataset 1polarities
use dataset 1representative citing papers
RACT is a retrieval-augmented self-supervised method that improves multi-table schema matching precision and completeness by up to 70% by probabilistically retrieving relevant tables to limit column candidate search space.
A vision for a cloud SmartNIC that hides Parquet decoding costs by offloading parsing and filters directly on the network datapath, backed by DuckDB performance estimates.
Presto is extended to GPU-aware execution using cuDF experiments on TPC-H, delivering up to 6x cost/performance gains over CPU Presto via optimized data paths and inter-operator communication.
AnnoRetrieve induces annotation schemas from documents and answers queries via structured retrieval, reporting F1 0.87 with lower LLM cost than graph/LLM baselines but higher cost than QUEST and vector search.
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.
-
RACT: Retrieval Augmented Column-Table Learning and Prediction for Multi-Table Schema Matching
RACT is a retrieval-augmented self-supervised method that improves multi-table schema matching precision and completeness by up to 70% by probabilistically retrieving relevant tables to limit column candidate search space.
-
Should I Hide My Duck in the Lake?
A vision for a cloud SmartNIC that hides Parquet decoding costs by offloading parsing and filters directly on the network datapath, backed by DuckDB performance estimates.
-
Accelerating Presto with GPUs
Presto is extended to GPU-aware execution using cuDF experiments on TPC-H, delivering up to 6x cost/performance gains over CPU Presto via optimized data paths and inter-operator communication.
-
AnnoRetrieve: Efficient Structured Retrieval for Unstructured Document Analysis
AnnoRetrieve induces annotation schemas from documents and answers queries via structured retrieval, reporting F1 0.87 with lower LLM cost than graph/LLM baselines but higher cost than QUEST and vector search.