@inproceedings{jang-etal-2026-ace,
title = "{ACE}: A Self-Correcting Agentic Canvas Editor for Multi-Slide Presentation Automation",
author = "Jang, JooYoung and
Lee, Taegyeong and
Park, Jihyeon and
Kwak, Nojun",
editor = "Atanasova, Pepa and
Haffari, Gholamreza and
Wang, Minghan",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-industry.74/",
pages = "1246--1271",
ISBN = "979-8-89176-460-6",
abstract = "Commercial design platforms increasingly edit documents through large language model (LLM) agents, but two practical problems block reliable deployment: legacy document formats expose only flat, absolutely positioned elements, so agents must recompute coordinates and routinely break layouts; and design has no unique ground truth, so diff-against-reference metrics penalize valid-but-different outputs. We present ACE, an agentic canvas editor over a hierarchical scene-graph with a presentation-specialized action space (98 tools), paired with CARE, a content-aware router that feeds the agent only the relevant slice of each deck (avg. $\sim 89\%$ input-token reduction), and a self-correction loop driven by a ground-truth-free instruction-following (IF) judge whose natural-language critique is fed back as the next-turn instruction. On a multi-slide benchmark and with a fixed backbone, a scene-graph editor in a single turn already matches a same-backbone agentic HTML pipeline that iterates internally; adding self-correction lifts ACE clearly above it on quality (IF 4.45 vs. 4.09; VQ 4.02 vs. 3.89) while remaining $1.75\times$ faster and $\sim 44\%$ cheaper. Self-correction contributes $+0.41$ IF / $+0.27$ VQ, and 66{\%} of cases halt after one pass, keeping it cheap."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="jang-etal-2026-ace">
<titleInfo>
<title>ACE: A Self-Correcting Agentic Canvas Editor for Multi-Slide Presentation Automation</title>
</titleInfo>
<name type="personal">
<namePart type="given">JooYoung</namePart>
<namePart type="family">Jang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Taegyeong</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jihyeon</namePart>
<namePart type="family">Park</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nojun</namePart>
<namePart type="family">Kwak</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>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track</title>
</titleInfo>
<name type="personal">
<namePart type="given">Pepa</namePart>
<namePart type="family">Atanasova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gholamreza</namePart>
<namePart type="family">Haffari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minghan</namePart>
<namePart type="family">Wang</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">979-8-89176-460-6</identifier>
</relatedItem>
<abstract>Commercial design platforms increasingly edit documents through large language model (LLM) agents, but two practical problems block reliable deployment: legacy document formats expose only flat, absolutely positioned elements, so agents must recompute coordinates and routinely break layouts; and design has no unique ground truth, so diff-against-reference metrics penalize valid-but-different outputs. We present ACE, an agentic canvas editor over a hierarchical scene-graph with a presentation-specialized action space (98 tools), paired with CARE, a content-aware router that feeds the agent only the relevant slice of each deck (avg. \sim 89% input-token reduction), and a self-correction loop driven by a ground-truth-free instruction-following (IF) judge whose natural-language critique is fed back as the next-turn instruction. On a multi-slide benchmark and with a fixed backbone, a scene-graph editor in a single turn already matches a same-backbone agentic HTML pipeline that iterates internally; adding self-correction lifts ACE clearly above it on quality (IF 4.45 vs. 4.09; VQ 4.02 vs. 3.89) while remaining 1.75\times faster and \sim 44% cheaper. Self-correction contributes +0.41 IF / +0.27 VQ, and 66% of cases halt after one pass, keeping it cheap.</abstract>
<identifier type="citekey">jang-etal-2026-ace</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-industry.74/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>1246</start>
<end>1271</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ACE: A Self-Correcting Agentic Canvas Editor for Multi-Slide Presentation Automation
%A Jang, JooYoung
%A Lee, Taegyeong
%A Park, Jihyeon
%A Kwak, Nojun
%Y Atanasova, Pepa
%Y Haffari, Gholamreza
%Y Wang, Minghan
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-460-6
%F jang-etal-2026-ace
%X Commercial design platforms increasingly edit documents through large language model (LLM) agents, but two practical problems block reliable deployment: legacy document formats expose only flat, absolutely positioned elements, so agents must recompute coordinates and routinely break layouts; and design has no unique ground truth, so diff-against-reference metrics penalize valid-but-different outputs. We present ACE, an agentic canvas editor over a hierarchical scene-graph with a presentation-specialized action space (98 tools), paired with CARE, a content-aware router that feeds the agent only the relevant slice of each deck (avg. \sim 89% input-token reduction), and a self-correction loop driven by a ground-truth-free instruction-following (IF) judge whose natural-language critique is fed back as the next-turn instruction. On a multi-slide benchmark and with a fixed backbone, a scene-graph editor in a single turn already matches a same-backbone agentic HTML pipeline that iterates internally; adding self-correction lifts ACE clearly above it on quality (IF 4.45 vs. 4.09; VQ 4.02 vs. 3.89) while remaining 1.75\times faster and \sim 44% cheaper. Self-correction contributes +0.41 IF / +0.27 VQ, and 66% of cases halt after one pass, keeping it cheap.
%U https://aclanthology.org/2026.emnlp-industry.74/
%P 1246-1271
Markdown (Informal)
[ACE: A Self-Correcting Agentic Canvas Editor for Multi-Slide Presentation Automation](https://aclanthology.org/2026.emnlp-industry.74/) (Jang et al., EMNLP 2026)
ACL