OctoSelector: Efficient and Effective Batch-Aware Model Selection for Large Language Models
Summary: OctoSelector learns difficulty-aware query representations to estimate multi-LLM accuracy, latency, and cost, then performs batch-aware selection via ILP under user objectives and constraints. It cuts NL2SQL cost by up to 67.7% while preserving accuracy and latency. (summarized by gpt-5.6-luna on Jul 26 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. GuangXue Zhang (University of California Irvine)
- 2. Yiming Lin (University of California Berkeley)
- 3. Sharad Mehrotra (University of California Irvine)
BibTeX Citation
@inproceedings{zhang_sigmod26,
title = {{OctoSelector: Efficient and Effective Batch-Aware Model Selection for Large Language Models}},
author = {Zhang, GuangXue and Lin, Yiming and Mehrotra, Sharad},
series = {{SIGMOD} '26},
booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
publisher = {Association for Computing Machinery},
doi = {10.1145/3802095},
url = {https://dl.acm.org/doi/10.1145/3802095},
year = {2026}
}
Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 10 of 10 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 10,504 | Task Cascades for Efficient Unstructured Data Processing | 2026 | SIGMOD |
| 2 | 279 | Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluation | 2024 | VLDB |
| 3 | 4,081 | Abacus: A Cost-Based Optimizer for Semantic Operator Systems | 2026 | VLDB |
| 4 | 7,809 | Can Large Language Models Be Query Optimizer for Relational Databases? | 2026 | SIGMOD |
| 5 | 10,733 | ScaleLLM: A Technique for Scalable LLM-augmented Data Systems | 2025 | SIGMOD |
| 6 | 10,483 | PRISM: Navigating Cost–Accuracy Trade-offs for NL2SQL | 2026 | SIGMOD |
| 7 | 10,510 | NL2SQLBench: A Modular Benchmarking Framework for LLM-Enabled NL2SQL Solutions | 2026 | VLDB |
| 8 | 8,829 | Cut Costs, Not Accuracy: LLM-Powered Data Processing with Guarantees | 2026 | SIGMOD |
| 9 | 10,856 | Optimized Batch Prompting for Cost-effective LLMs | 2025 | VLDB |
| 10 | 8,627 | ThriftLLM: On Cost-Effective Selection of Large Language Models for Classification Queries | 2025 | VLDB |