@inproceedings{hayat-etal-2026-le,
title = "{LE}-Viz: Complementing Text Serialization with Visual Encoding of Structured Temporal Data for Cross-Distribution Generalization",
author = "Hayat, Ahatsham and
Khan, Bilal and
Hasan, Mohammad Rashedul",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.29/",
pages = "578--599",
ISBN = "979-8-89176-462-0",
abstract = "When large language models process structured temporal data, serialization into token sequences collapses the two-dimensional organization ($\textit{Features * Time}$) that encodes meaningful relationships between co-occurring features. We investigate whether visual encoding can recover the structural information that linearization loses. We introduce LE-Viz, a framework that transforms longitudinal experiential data into complementary visual and textual representations for a vision-language model that generates prospective narratives, framing prediction as language generation. Evaluated on cross-distribution tasks across three datasets spanning mental health and educational forecasting, LE-Viz sets a new benchmark on a widely-adopted cross-distribution generalization task, improving over both the traditional state-of-the-art and the best text-only LLM baseline, and consistently outperforms all baselines. Analyses show that spatial organization of visual inputs, not information quantity, drives the generalization gain, and that the multimodal combination exceeds both text-only and vision-only conditions."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hayat-etal-2026-le">
<titleInfo>
<title>LE-Viz: Complementing Text Serialization with Visual Encoding of Structured Temporal Data for Cross-Distribution Generalization</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ahatsham</namePart>
<namePart type="family">Hayat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bilal</namePart>
<namePart type="family">Khan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Rashedul</namePart>
<namePart type="family">Hasan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>When large language models process structured temporal data, serialization into token sequences collapses the two-dimensional organization (Features * Time) that encodes meaningful relationships between co-occurring features. We investigate whether visual encoding can recover the structural information that linearization loses. We introduce LE-Viz, a framework that transforms longitudinal experiential data into complementary visual and textual representations for a vision-language model that generates prospective narratives, framing prediction as language generation. Evaluated on cross-distribution tasks across three datasets spanning mental health and educational forecasting, LE-Viz sets a new benchmark on a widely-adopted cross-distribution generalization task, improving over both the traditional state-of-the-art and the best text-only LLM baseline, and consistently outperforms all baselines. Analyses show that spatial organization of visual inputs, not information quantity, drives the generalization gain, and that the multimodal combination exceeds both text-only and vision-only conditions.</abstract>
<identifier type="citekey">hayat-etal-2026-le</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.29/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>578</start>
<end>599</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LE-Viz: Complementing Text Serialization with Visual Encoding of Structured Temporal Data for Cross-Distribution Generalization
%A Hayat, Ahatsham
%A Khan, Bilal
%A Hasan, Mohammad Rashedul
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F hayat-etal-2026-le
%X When large language models process structured temporal data, serialization into token sequences collapses the two-dimensional organization (Features * Time) that encodes meaningful relationships between co-occurring features. We investigate whether visual encoding can recover the structural information that linearization loses. We introduce LE-Viz, a framework that transforms longitudinal experiential data into complementary visual and textual representations for a vision-language model that generates prospective narratives, framing prediction as language generation. Evaluated on cross-distribution tasks across three datasets spanning mental health and educational forecasting, LE-Viz sets a new benchmark on a widely-adopted cross-distribution generalization task, improving over both the traditional state-of-the-art and the best text-only LLM baseline, and consistently outperforms all baselines. Analyses show that spatial organization of visual inputs, not information quantity, drives the generalization gain, and that the multimodal combination exceeds both text-only and vision-only conditions.
%U https://aclanthology.org/2026.findings-emnlp.29/
%P 578-599
Markdown (Informal)
[LE-Viz: Complementing Text Serialization with Visual Encoding of Structured Temporal Data for Cross-Distribution Generalization](https://aclanthology.org/2026.findings-emnlp.29/) (Hayat et al., Findings 2026)
ACL