LMRPA: Large Language Model-Driven Efficient Robotic Process Automation for OCR

Osama Hosam Abdellaif, Abdelrahman Nader Hassan, Ali Hamdi · 2024

Read LMRPA: Large Language Model-Driven Efficient Robotic Process Automation for OCR by Osama Hosam Abdellaif, Abdelrahman Nader Hassan, Ali Hamdi (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-18063.txt.

Description

Osama Hosam Abdellaif and co-authors speed up document automation by pairing it with large language models in a 2024 paper. Robotic process automation handles repetitive office tasks, but reading text from images, called OCR, often slows it down. Errors on unclear characters make the bottleneck worse.

Their system, LMRPA, uses a language model to clean up and sharpen the text that OCR extracts. Benchmarks against two leading automation platforms, with two OCR engines, showed large time savings. In one test batch the system finished in under ten seconds where rivals took over eighteen. Gains of up to fifty-two percent were reported across tasks. The authors present the pairing as a practical upgrade for high-volume document work.

Catalog record

CREATOROsama Hosam Abdellaif, Abdelrahman Nader Hassan, Ali Hamdi
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE21 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-18063

Browse the full library · About the Tower of Babel