Presentations are not always linear! GNN meets LLM for Document-to-Presentation Transformation with Attribution
Automatically generating a presentation from the text of a long document is a challenging and useful problem. In contrast to a flat summary, a presentation needs to have a better and non-linear narrative, i.e., the content of a slide can come from different and non-contiguous parts of the given docu...
Gespeichert in:
Hauptverfasser: | , , , |
---|---|
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | |
container_volume | |
creator | Maheshwari, Himanshu Bandyopadhyay, Sambaran Garimella, Aparna Natarajan, Anandhavelu |
description | Automatically generating a presentation from the text of a long document is a
challenging and useful problem. In contrast to a flat summary, a presentation
needs to have a better and non-linear narrative, i.e., the content of a slide
can come from different and non-contiguous parts of the given document.
However, it is difficult to incorporate such non-linear mapping of content to
slides and ensure that the content is faithful to the document. LLMs are prone
to hallucination and their performance degrades with the length of the input
document. Towards this, we propose a novel graph based solution where we learn
a graph from the input document and use a combination of graph neural network
and LLM to generate a presentation with attribution of content for each slide.
We conduct thorough experiments to show the merit of our approach compared to
directly using LLMs for this task. |
doi_str_mv | 10.48550/arxiv.2405.13095 |
format | Article |
fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2405_13095</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2405_13095</sourcerecordid><originalsourceid>FETCH-LOGICAL-a675-b40ed60ad04346067f36297523df6ab93b7e68a5955b80ebce784e3ea55fd39c3</originalsourceid><addsrcrecordid>eNpNj8tOwzAURL1hgQofwIrbD0hw40eSZVWgIIXSRfbRdXIjLCUxsl1K_56-FqxGM5oZ6TD2sOCpLJTiT-h_7U-aSa7SheClumVu6ynQFDFaNwVATzC5CDjs8RBgsBOhn8N6s4GRKAaoqg_onYdn1-7G4y6JLvl_AbXHKRwb48XubfyCZYzemt0puGM3PQ6B7q86Y_XrS716S6rP9ftqWSWoc5UYyanTHDsuhdRc573QWZmrTHS9RlMKk5MuUJVKmYKTaSkvJAlCpfpOlK2YscfL7Zm4-fZ2RH9oTuTNmVz8AaKVVXU</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Presentations are not always linear! GNN meets LLM for Document-to-Presentation Transformation with Attribution</title><source>arXiv.org</source><creator>Maheshwari, Himanshu ; Bandyopadhyay, Sambaran ; Garimella, Aparna ; Natarajan, Anandhavelu</creator><creatorcontrib>Maheshwari, Himanshu ; Bandyopadhyay, Sambaran ; Garimella, Aparna ; Natarajan, Anandhavelu</creatorcontrib><description>Automatically generating a presentation from the text of a long document is a
challenging and useful problem. In contrast to a flat summary, a presentation
needs to have a better and non-linear narrative, i.e., the content of a slide
can come from different and non-contiguous parts of the given document.
However, it is difficult to incorporate such non-linear mapping of content to
slides and ensure that the content is faithful to the document. LLMs are prone
to hallucination and their performance degrades with the length of the input
document. Towards this, we propose a novel graph based solution where we learn
a graph from the input document and use a combination of graph neural network
and LLM to generate a presentation with attribution of content for each slide.
We conduct thorough experiments to show the merit of our approach compared to
directly using LLMs for this task.</description><identifier>DOI: 10.48550/arxiv.2405.13095</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Computation and Language</subject><creationdate>2024-05</creationdate><rights>http://creativecommons.org/licenses/by-nc-sa/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2405.13095$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2405.13095$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Maheshwari, Himanshu</creatorcontrib><creatorcontrib>Bandyopadhyay, Sambaran</creatorcontrib><creatorcontrib>Garimella, Aparna</creatorcontrib><creatorcontrib>Natarajan, Anandhavelu</creatorcontrib><title>Presentations are not always linear! GNN meets LLM for Document-to-Presentation Transformation with Attribution</title><description>Automatically generating a presentation from the text of a long document is a
challenging and useful problem. In contrast to a flat summary, a presentation
needs to have a better and non-linear narrative, i.e., the content of a slide
can come from different and non-contiguous parts of the given document.
However, it is difficult to incorporate such non-linear mapping of content to
slides and ensure that the content is faithful to the document. LLMs are prone
to hallucination and their performance degrades with the length of the input
document. Towards this, we propose a novel graph based solution where we learn
a graph from the input document and use a combination of graph neural network
and LLM to generate a presentation with attribution of content for each slide.
We conduct thorough experiments to show the merit of our approach compared to
directly using LLMs for this task.</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Computation and Language</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNpNj8tOwzAURL1hgQofwIrbD0hw40eSZVWgIIXSRfbRdXIjLCUxsl1K_56-FqxGM5oZ6TD2sOCpLJTiT-h_7U-aSa7SheClumVu6ynQFDFaNwVATzC5CDjs8RBgsBOhn8N6s4GRKAaoqg_onYdn1-7G4y6JLvl_AbXHKRwb48XubfyCZYzemt0puGM3PQ6B7q86Y_XrS716S6rP9ftqWSWoc5UYyanTHDsuhdRc573QWZmrTHS9RlMKk5MuUJVKmYKTaSkvJAlCpfpOlK2YscfL7Zm4-fZ2RH9oTuTNmVz8AaKVVXU</recordid><startdate>20240521</startdate><enddate>20240521</enddate><creator>Maheshwari, Himanshu</creator><creator>Bandyopadhyay, Sambaran</creator><creator>Garimella, Aparna</creator><creator>Natarajan, Anandhavelu</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20240521</creationdate><title>Presentations are not always linear! GNN meets LLM for Document-to-Presentation Transformation with Attribution</title><author>Maheshwari, Himanshu ; Bandyopadhyay, Sambaran ; Garimella, Aparna ; Natarajan, Anandhavelu</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a675-b40ed60ad04346067f36297523df6ab93b7e68a5955b80ebce784e3ea55fd39c3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Computation and Language</topic><toplevel>online_resources</toplevel><creatorcontrib>Maheshwari, Himanshu</creatorcontrib><creatorcontrib>Bandyopadhyay, Sambaran</creatorcontrib><creatorcontrib>Garimella, Aparna</creatorcontrib><creatorcontrib>Natarajan, Anandhavelu</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Maheshwari, Himanshu</au><au>Bandyopadhyay, Sambaran</au><au>Garimella, Aparna</au><au>Natarajan, Anandhavelu</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Presentations are not always linear! GNN meets LLM for Document-to-Presentation Transformation with Attribution</atitle><date>2024-05-21</date><risdate>2024</risdate><abstract>Automatically generating a presentation from the text of a long document is a
challenging and useful problem. In contrast to a flat summary, a presentation
needs to have a better and non-linear narrative, i.e., the content of a slide
can come from different and non-contiguous parts of the given document.
However, it is difficult to incorporate such non-linear mapping of content to
slides and ensure that the content is faithful to the document. LLMs are prone
to hallucination and their performance degrades with the length of the input
document. Towards this, we propose a novel graph based solution where we learn
a graph from the input document and use a combination of graph neural network
and LLM to generate a presentation with attribution of content for each slide.
We conduct thorough experiments to show the merit of our approach compared to
directly using LLMs for this task.</abstract><doi>10.48550/arxiv.2405.13095</doi><oa>free_for_read</oa></addata></record> |
fulltext | fulltext_linktorsrc |
identifier | DOI: 10.48550/arxiv.2405.13095 |
ispartof | |
issn | |
language | eng |
recordid | cdi_arxiv_primary_2405_13095 |
source | arXiv.org |
subjects | Computer Science - Artificial Intelligence Computer Science - Computation and Language |
title | Presentations are not always linear! GNN meets LLM for Document-to-Presentation Transformation with Attribution |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-28T23%3A36%3A25IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Presentations%20are%20not%20always%20linear!%20GNN%20meets%20LLM%20for%20Document-to-Presentation%20Transformation%20with%20Attribution&rft.au=Maheshwari,%20Himanshu&rft.date=2024-05-21&rft_id=info:doi/10.48550/arxiv.2405.13095&rft_dat=%3Carxiv_GOX%3E2405_13095%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |