A fine-tuned Llama3-8B model generates receipt layouts with fewer overlapping boxes than LayoutTransformer, and adding bounding box coordinates to text raises BERT classification accuracy from 62% to 74%.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Spatial Information Integration in Small Language Models for Document Layout Generation and Classification
A fine-tuned Llama3-8B model generates receipt layouts with fewer overlapping boxes than LayoutTransformer, and adding bounding box coordinates to text raises BERT classification accuracy from 62% to 74%.