Research paper

CACHE-ED2: Compiling LLM Reasoning into Reusable Extraction Programs for Document Extraction at Scale

Research from AmazonPublished by Association for Computational LinguisticsOct 2026PDF

Explore a method that turns repeated language-model reasoning into reusable programs for document extraction.

CACHE-ED2: Compiling LLM Reasoning into Reusable Extraction Programs for Document Extraction at Scale cover

Resource access

Get the full resource

Enter your business details to read the full resource.

First nameLast nameWork emailCompanyJob titleCountry or region

Sudhanshu Bhoi, Anurag Tripathi, Amir Raza, Mayank Jauhari. CACHE-ED2: Compiling LLM Reasoning into Reusable Extraction Programs for Document Extraction at Scale. Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track. © 2026 Association for Computational Linguistics. CC BY 4.0. 10.18653/2026.emnlp-industry.93. Original PDF, unmodified.

CC BY 4.0