DocETL: Agentic Query Rewriting and Evaluation for Complex Document Processing
Summary: DocETL declaratively optimizes complex LLM document-processing pipelines via agent-generated logical rewrites (“rewrite directives”) and latency-aware plan evaluation. It trades single-call execution for decomposition and empirical plan search, improving accuracy 21–80% on four real tasks. (summarized by gpt-5.6-luna on Jul 24 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Shreya Shankar (University of California Berkeley)
- 2. Tristan Chambers (Berkeley Institute for Data Science)
- 3. Tarak Shah (Berkeley Institute for Data Science)
- 4. Aditya G. Parameswaran (University of California Berkeley)
- 5. Eugene Wu (Columbia University)
BibTeX Citation
@article{shankar_vldb25,
title = {{DocETL: Agentic Query Rewriting and Evaluation for Complex Document Processing}},
author = {Shankar, Shreya and Chambers, Tristan and Shah, Tarak and Parameswaran, Aditya G. and Wu, Eugene},
journal = {PVLDB},
series = {{VLDB} '25},
volume = {18},
number = {9},
pages = {3035--3048},
doi = {10.14778/3746405.3746426},
url = {https://doi.org/10.14778/3746405.3746426},
year = {2025}
}
Incoming Citations (Sorted by Pagerank)
Showing 26 of 26 citing papers.
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 18 of 18 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 | 10,286 | ScaleDoc: Scaling LLM-based Predicates over Large Document Collections | 2026 | SIGMOD |
| 2 | 6,371 | QUEST: Query Optimization in Unstructured Document Analysis | 2025 | VLDB |
| 3 | 7,439 | AOP: Automated and Interactive LLM Pipeline Orchestration for Answering Complex Queries | 2025 | CIDR |
| 4 | 10,190 | AgenticScholar: Agentic Data Management with Pipeline Orchestration for Scholarly Corpora | 2026 | SIGMOD |
| 5 | 10,614 | LLM-AutoDP: Automatic Data Processing via LLM Agents for Model Fine-tuning | 2026 | VLDB |
| 6 | 10,719 | Doctopus: A System for Budget-aware Structural Data Extraction from Unstructured Documents | 2025 | SIGMOD |
| 7 | 11,186 | Unstructured Data Fusion for Schema and Data Extraction | 2024 | SIGMOD |
| 8 | 10,504 | Task Cascades for Efficient Unstructured Data Processing | 2026 | SIGMOD |
| 9 | 8,340 | Doctopus: Budget-aware Structural Table Extraction from Unstructured Documents | 2025 | VLDB |
| 10 | 13,339 | DocDB: A Database for Unstructured Document Analysis | 2025 | VLDB |