Structsum Generation for Faster Text Comprehension

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jain, Parag, Marzoca, Andreea, Piccinno, Francesco
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910494871257088
author Jain, Parag
Marzoca, Andreea
Piccinno, Francesco
author_facet Jain, Parag
Marzoca, Andreea
Piccinno, Francesco
contents We consider the task of generating structured representations of text using large language models (LLMs). We focus on tables and mind maps as representative modalities. Tables are more organized way of representing data, while mind maps provide a visually dynamic and flexible approach, particularly suitable for sparse content. Despite the effectiveness of LLMs on different tasks, we show that current models struggle with generating structured outputs. In response, we present effective prompting strategies for both of these tasks. We introduce a taxonomy of problems around factuality, global and local structure, common to both modalities and propose a set of critiques to tackle these issues resulting in an absolute improvement in accuracy of +37pp (79%) for mind maps and +15pp (78%) for tables. To evaluate semantic coverage of generated structured representations we propose Auto-QA, and we verify the adequacy of Auto-QA using SQuAD dataset. We further evaluate the usefulness of structured representations via a text comprehension user study. The results show a significant reduction in comprehension time compared to text when using table (42.9%) and mind map (31.9%), without loss in accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2401_06837
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Structsum Generation for Faster Text Comprehension
Jain, Parag
Marzoca, Andreea
Piccinno, Francesco
Computation and Language
Artificial Intelligence
We consider the task of generating structured representations of text using large language models (LLMs). We focus on tables and mind maps as representative modalities. Tables are more organized way of representing data, while mind maps provide a visually dynamic and flexible approach, particularly suitable for sparse content. Despite the effectiveness of LLMs on different tasks, we show that current models struggle with generating structured outputs. In response, we present effective prompting strategies for both of these tasks. We introduce a taxonomy of problems around factuality, global and local structure, common to both modalities and propose a set of critiques to tackle these issues resulting in an absolute improvement in accuracy of +37pp (79%) for mind maps and +15pp (78%) for tables. To evaluate semantic coverage of generated structured representations we propose Auto-QA, and we verify the adequacy of Auto-QA using SQuAD dataset. We further evaluate the usefulness of structured representations via a text comprehension user study. The results show a significant reduction in comprehension time compared to text when using table (42.9%) and mind map (31.9%), without loss in accuracy.
title Structsum Generation for Faster Text Comprehension
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2401.06837