Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2603.11622 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915856822304768 |
|---|---|
| author | Qi, Kangkang Xie, Dongyang Li, Wenbo Zhang, Hao Zhu, Yuanyuan Yu, Jeffrey Xu Zhao, Kangfei |
| author_facet | Qi, Kangkang Xie, Dongyang Li, Wenbo Zhang, Hao Zhu, Yuanyuan Yu, Jeffrey Xu Zhao, Kangfei |
| contents | The integration of Large Language Models (LLMs) into data analytics has unlocked powerful capabilities for reasoning over bulk structured and unstructured data. However, existing systems typically rely on either DataFrame primitives, which lack the efficient execution infrastructure of modern DBMSs, or SQL User-Defined Functions (UDFs), which isolate semantic logic from the query optimizer and burden users with implementation complexities. The LLM-powered semantic operators also bring new challenges due to the high cost and non-deterministic nature of LLM invocation, where conventional optimization rules and cost models are inapplicable for their optimization.
To bridge these gaps, we present Sema, a high-performance semantic query engine built on DuckDB that treats LLM-powered semantic operators as first-class citizens. Sema introduces SemaSQL, a declarative dialect that allows users seamlessly inject natural language expressions into standard SQL clauses, enabling end-to-end optimization and execution. At the logical level, the optimizer of Sema compresses natural language expressions and deduces relational constraints from semantic operators. At runtime, Sema employs Adaptive Query Execution (AQE) to dynamically reorder operators, fuse semantic operations, and apply prompt batching. This approach seeks a Pareto-optimal execution path balancing token consumption and latency under accuracy constraints. We evaluate Sema on 20 semantic queries across classification, summarization, and extraction tasks. Experimental results demonstrate that Sema achieves $2-10 \times$ speedup against three baseline systems while achieving competitive result quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_11622 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Sema: A High-performance System for LLM-based Semantic Query Processing Qi, Kangkang Xie, Dongyang Li, Wenbo Zhang, Hao Zhu, Yuanyuan Yu, Jeffrey Xu Zhao, Kangfei Databases The integration of Large Language Models (LLMs) into data analytics has unlocked powerful capabilities for reasoning over bulk structured and unstructured data. However, existing systems typically rely on either DataFrame primitives, which lack the efficient execution infrastructure of modern DBMSs, or SQL User-Defined Functions (UDFs), which isolate semantic logic from the query optimizer and burden users with implementation complexities. The LLM-powered semantic operators also bring new challenges due to the high cost and non-deterministic nature of LLM invocation, where conventional optimization rules and cost models are inapplicable for their optimization. To bridge these gaps, we present Sema, a high-performance semantic query engine built on DuckDB that treats LLM-powered semantic operators as first-class citizens. Sema introduces SemaSQL, a declarative dialect that allows users seamlessly inject natural language expressions into standard SQL clauses, enabling end-to-end optimization and execution. At the logical level, the optimizer of Sema compresses natural language expressions and deduces relational constraints from semantic operators. At runtime, Sema employs Adaptive Query Execution (AQE) to dynamically reorder operators, fuse semantic operations, and apply prompt batching. This approach seeks a Pareto-optimal execution path balancing token consumption and latency under accuracy constraints. We evaluate Sema on 20 semantic queries across classification, summarization, and extraction tasks. Experimental results demonstrate that Sema achieves $2-10 \times$ speedup against three baseline systems while achieving competitive result quality. |
| title | Sema: A High-performance System for LLM-based Semantic Query Processing |
| topic | Databases |
| url | https://arxiv.org/abs/2603.11622 |