Access Paths for Efficient Ordering with Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhao, Fuheng, Chen, Jiayue, Pan, Yiming, Rabbani, Tahseen, Sohaib, Agrawal, Divyakant, Abbadi, Amr El, Aggarwal, Paritosh, Datta, Anupam, Tsirogiannis, Dimitris
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916029565763584
author Zhao, Fuheng
Chen, Jiayue
Pan, Yiming
Rabbani, Tahseen
Sohaib
Agrawal, Divyakant
Abbadi, Amr El
Aggarwal, Paritosh
Datta, Anupam
Tsirogiannis, Dimitris
author_facet Zhao, Fuheng
Chen, Jiayue
Pan, Yiming
Rabbani, Tahseen
Sohaib
Agrawal, Divyakant
Abbadi, Amr El
Aggarwal, Paritosh
Datta, Anupam
Tsirogiannis, Dimitris
contents In this work, we present the \texttt{LLM ORDER BY} semantic operator as a logical abstraction and conduct a systematic study of its physical implementations. First, we propose several improvements to existing semantic sorting algorithms and introduce a semantic-aware external merge sort algorithm. Our extensive evaluation reveals that no single implementation offers universal optimality on all datasets. From our evaluations, we observe a general test-time scaling relationship between sorting cost and the ordering quality for comparison-based algorithms. Building on these insights, we design a budget-aware optimizer that utilizes heuristic rules, LLM-as-Judge evaluation, and consensus aggregation to dynamically select the near-optimal access path for LLM ORDER BY. In our extensive evaluations, our optimizer consistently achieves ranking accuracy on par with or superior to the best static methods across all benchmarks. We believe that this work provides foundational insights into the principled optimization of semantic operators essential for building robust, large-scale LLM-powered analytic systems.
format Preprint
id arxiv_https___arxiv_org_abs_2509_00303
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Access Paths for Efficient Ordering with Large Language Models
Zhao, Fuheng
Chen, Jiayue
Pan, Yiming
Rabbani, Tahseen
Sohaib
Agrawal, Divyakant
Abbadi, Amr El
Aggarwal, Paritosh
Datta, Anupam
Tsirogiannis, Dimitris
Databases
Artificial Intelligence
Information Retrieval
In this work, we present the \texttt{LLM ORDER BY} semantic operator as a logical abstraction and conduct a systematic study of its physical implementations. First, we propose several improvements to existing semantic sorting algorithms and introduce a semantic-aware external merge sort algorithm. Our extensive evaluation reveals that no single implementation offers universal optimality on all datasets. From our evaluations, we observe a general test-time scaling relationship between sorting cost and the ordering quality for comparison-based algorithms. Building on these insights, we design a budget-aware optimizer that utilizes heuristic rules, LLM-as-Judge evaluation, and consensus aggregation to dynamically select the near-optimal access path for LLM ORDER BY. In our extensive evaluations, our optimizer consistently achieves ranking accuracy on par with or superior to the best static methods across all benchmarks. We believe that this work provides foundational insights into the principled optimization of semantic operators essential for building robust, large-scale LLM-powered analytic systems.
title Access Paths for Efficient Ordering with Large Language Models
topic Databases
Artificial Intelligence
Information Retrieval
url https://arxiv.org/abs/2509.00303