Abacus: A Cost-Based Optimizer for Semantic Operator Systems
Summary: Abacus: cost-based optimizer for semantic-operator pipelines over unstructured docs, choosing physical implementations of LLM maps/filters/joins under quality–cost–latency objectives. Uses few validation examples / priors / LLM judge to estimate operator performance and globally optimize end-to-end systems. (summarized by gpt-5.4-mini on Apr 12 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Matthew Russo (Massachusetts Institute of Technology)
- 2. Chunwei Liu (Massachusetts Institute of Technology)
- 3. Sivaprasad Sudhir (Massachusetts Institute of Technology)
- 4. Gerardo Vitagliano (Massachusetts Institute of Technology)
- 5. Michael Cafarella (Massachusetts Institute of Technology)
- 6. Tim Kraska (Massachusetts Institute of Technology)
- 7. Samuel Madden (Massachusetts Institute of Technology)
BibTeX Citation
@article{russo_vldb26,
title = {{Abacus: A Cost-Based Optimizer for Semantic Operator Systems}},
author = {Russo, Matthew and Liu, Chunwei and Sudhir, Sivaprasad and Vitagliano, Gerardo and Cafarella, Michael and Kraska, Tim and Madden, Samuel},
journal = {PVLDB},
series = {{VLDB} '26},
volume = {19},
number = {5},
pages = {1060--1073},
doi = {10.14778/3796195.3796215},
url = {https://doi.org/10.14778/3796195.3796215},
year = {2026}
}Incoming Citations (Sorted by Pagerank)
Showing 17 of 17 citing papers.
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 15 of 15 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next