SAGE: A Unified Algebra and Self-Adaptive Execution for AI Functions in SQL

arXiv:2608.20630 · cs.AI · Submitted 2026-08-21 · Read on arXiv

cs.AI

Submitted: 2026-08-21

Updated: 2026-08-21

Comments: arXiv admin comment: This version has been removed by arXiv administrators as the submitter did not have the rights to agree to the license at the time of submission

Code: https://github.com/BerriAI/litel

License: http://creativecommons.org/licenses/by-nc-nd/4.0/

The gist: SQL systems increasingly expose AI functions for tasks such as classification, extraction, filtering, ranking, retrieval, joining, and summarization.

Terminology

Abstract

SQL systems increasingly expose AI functions for tasks such as classification, extraction, filtering, ranking, retrieval, joining, and summarization. Despite their diverse APIs, these functions play only three relational roles: transforming individual rows, aggregating groups, or generating relationships between row pairs. We present SAGE (Self-Adaptive Generative Execution), a unified logical and physical framework that captures these roles with three typed primitives, AI SCALAR, AI AGG, and AI JOIN, and composes them naturally with standard relational operators. All primitives share a confidence-gated execution interface while supporting physical strategies tailored to their relational shape. The main challenge is AI JOIN, where SAGE analyzes the predicate, decomposes compound conditions when possible, and uses a recipe card together with a small label-free probe to select among complete execution strategies. Across a broad audit of public AI operators and evaluations spanning scalar, aggregate, and join workloads, this formulation covers common AI functionality while consistently improving execution quality and efficiency. SAGE achieves the strongest overall SemBench performance and, on a representative factorable join, reduces pairwise model calls by more than two orders of magnitude, yielding a 358-fold measured cost reduction.

Sources

Related papers