Saved in:
Bibliographic Details
Main Authors: Rückstieß, Thomas, Huang, Alana, Vujanic, Robin
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.17348
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912166641139712
author Rückstieß, Thomas
Huang, Alana
Vujanic, Robin
author_facet Rückstieß, Thomas
Huang, Alana
Vujanic, Robin
contents Despite the popularity and widespread use of semi-structured data formats such as JSON, end-to-end supervised learning applied directly to such data remains underexplored. We present ORIGAMI (Object RepresentatIon via Generative Autoregressive ModellIng), a transformer-based architecture that directly processes nested key/value pairs while preserving their hierarchical semantics. Our key technical contributions include: (1) a structure-preserving tokenizer, (2) a novel key/value position encoding scheme, and (3) a grammar-constrained training and inference framework that ensures valid outputs and accelerates training convergence. These enhancements enable efficient end-to-end modeling of semi-structured data. By reformulating classification as next-token prediction, ORIGAMI naturally handles both single-label and multi-label tasks without architectural modifications. Empirical evaluation across diverse domains demonstrates ORIGAMI's effectiveness: On standard tabular benchmarks converted to JSON, ORIGAMI remains competitive with classical and state-of-the-art approaches. On native JSON datasets, we outperform baselines on multi-label classification and specialized models such as convolutional and graph neural networks on a code classification task. Through extensive ablation studies, we validate the impact of each architectural component and establish ORIGAMI as a robust framework for end-to-end learning on semi-structured data.
format Preprint
id arxiv_https___arxiv_org_abs_2412_17348
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ORIGAMI: A generative transformer architecture for predictions from semi-structured data
Rückstieß, Thomas
Huang, Alana
Vujanic, Robin
Machine Learning
Despite the popularity and widespread use of semi-structured data formats such as JSON, end-to-end supervised learning applied directly to such data remains underexplored. We present ORIGAMI (Object RepresentatIon via Generative Autoregressive ModellIng), a transformer-based architecture that directly processes nested key/value pairs while preserving their hierarchical semantics. Our key technical contributions include: (1) a structure-preserving tokenizer, (2) a novel key/value position encoding scheme, and (3) a grammar-constrained training and inference framework that ensures valid outputs and accelerates training convergence. These enhancements enable efficient end-to-end modeling of semi-structured data. By reformulating classification as next-token prediction, ORIGAMI naturally handles both single-label and multi-label tasks without architectural modifications. Empirical evaluation across diverse domains demonstrates ORIGAMI's effectiveness: On standard tabular benchmarks converted to JSON, ORIGAMI remains competitive with classical and state-of-the-art approaches. On native JSON datasets, we outperform baselines on multi-label classification and specialized models such as convolutional and graph neural networks on a code classification task. Through extensive ablation studies, we validate the impact of each architectural component and establish ORIGAMI as a robust framework for end-to-end learning on semi-structured data.
title ORIGAMI: A generative transformer architecture for predictions from semi-structured data
topic Machine Learning
url https://arxiv.org/abs/2412.17348