OmniStruct: Universal Text-to-Structure Generation across Diverse Schemas

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Huang, James Y., Zhou, Wenxuan, Xu, Nan, Wang, Fei, Liu, Qin, Zhang, Sheng, Poon, Hoifung, Chen, Muhao
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917099904958464
author Huang, James Y.
Zhou, Wenxuan
Xu, Nan
Wang, Fei
Liu, Qin
Zhang, Sheng
Poon, Hoifung
Chen, Muhao
author_facet Huang, James Y.
Zhou, Wenxuan
Xu, Nan
Wang, Fei
Liu, Qin
Zhang, Sheng
Poon, Hoifung
Chen, Muhao
contents The ability of Large Language Models (LLMs) to generate structured outputs that follow arbitrary schemas is crucial to a wide range of downstream tasks that require diverse structured representations of results such as information extraction, table generation, and function calling. While modern LLMs excel in generating unstructured responses in natural language, whether this advancement translates to a strong performance on text-to-structure tasks remains unclear. To bridge this gap, we first introduce OmniStruct, a comprehensive benchmark for assessing LLMs' capabilities on diverse text-to-structure tasks such as information extraction, table generation, and function calling. We build OmniStruct by identifying existing datasets across a wide range of tasks that are suitable for a structured answer format, and adapting them under a unified text-to-structure problem setting. To facilitate the development of efficient text-to-structure models, we collect high-quality training data via synthetic task generation. Without using any supervised data for OmniStruct tasks, our experiments demonstrate the possibility of fine-tuning much smaller models on synthetic data into universal structured generation models that can rival the performance of GPT-4o.
format Preprint
id arxiv_https___arxiv_org_abs_2511_18335
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle OmniStruct: Universal Text-to-Structure Generation across Diverse Schemas
Huang, James Y.
Zhou, Wenxuan
Xu, Nan
Wang, Fei
Liu, Qin
Zhang, Sheng
Poon, Hoifung
Chen, Muhao
Computation and Language
Artificial Intelligence
Machine Learning
The ability of Large Language Models (LLMs) to generate structured outputs that follow arbitrary schemas is crucial to a wide range of downstream tasks that require diverse structured representations of results such as information extraction, table generation, and function calling. While modern LLMs excel in generating unstructured responses in natural language, whether this advancement translates to a strong performance on text-to-structure tasks remains unclear. To bridge this gap, we first introduce OmniStruct, a comprehensive benchmark for assessing LLMs' capabilities on diverse text-to-structure tasks such as information extraction, table generation, and function calling. We build OmniStruct by identifying existing datasets across a wide range of tasks that are suitable for a structured answer format, and adapting them under a unified text-to-structure problem setting. To facilitate the development of efficient text-to-structure models, we collect high-quality training data via synthetic task generation. Without using any supervised data for OmniStruct tasks, our experiments demonstrate the possibility of fine-tuning much smaller models on synthetic data into universal structured generation models that can rival the performance of GPT-4o.
title OmniStruct: Universal Text-to-Structure Generation across Diverse Schemas
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2511.18335