ORIGAMI is a generative transformer with structure-preserving tokenization, key/value position encoding, and grammar-constrained decoding that matches or beats baselines on tabular, multi-label, and code-classification tasks.
Large language models (LLMs) on tabular data: Prediction, generation, and understanding - a survey
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
ORIGAMI: A generative transformer architecture for predictions from semi-structured data
ORIGAMI is a generative transformer with structure-preserving tokenization, key/value position encoding, and grammar-constrained decoding that matches or beats baselines on tabular, multi-label, and code-classification tasks.