@inproceedings{xu-etal-2026-cognitive,
title = "The Cognitive Bandwidth Bottleneck: Shifting Long-Horizon Agents from Planning with Actions to Planning with Schemas",
author = "Xu, Baixuan and
Zheng, Tianshi and
Wang, Zhaowei and
Tsang, Hong Ting and
Wang, Weiqi and
Fang, Tianqing and
Song, Yangqiu",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.755/",
pages = "17451--17468",
ISBN = "979-8-89176-459-0",
abstract = "Long-horizon LLM agents typically choose from an environment-enumerated list of grounded actions (Planning with Actions, PwA). While this interface reduces formatting risk, scaling the list increases screening load and can bottleneck planning. We study Planning with Schemas (PwS), which replaces action enumeration with a small set of action templates that the agent instantiates into executable actions. We introduce a cognitive-bandwidth perspective of this trade-off and quantify screening vs. instantiation pressure using context length proxy and failure actions proxy. Across four interactive environments, we identify an inflection point: PwA excels at small interfaces but degrades as action lists grow, whereas PwS becomes increasingly competitive and often superior. A distractor-injection stress test reproduces the PwA drop by enlarging the action space alone, isolating screening load and highlighting schema instantiation as the key remaining bottleneck."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="xu-etal-2026-cognitive">
<titleInfo>
<title>The Cognitive Bandwidth Bottleneck: Shifting Long-Horizon Agents from Planning with Actions to Planning with Schemas</title>
</titleInfo>
<name type="personal">
<namePart type="given">Baixuan</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tianshi</namePart>
<namePart type="family">Zheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhaowei</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hong</namePart>
<namePart type="given">Ting</namePart>
<namePart type="family">Tsang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weiqi</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tianqing</namePart>
<namePart type="family">Fang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yangqiu</namePart>
<namePart type="family">Song</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</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">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Long-horizon LLM agents typically choose from an environment-enumerated list of grounded actions (Planning with Actions, PwA). While this interface reduces formatting risk, scaling the list increases screening load and can bottleneck planning. We study Planning with Schemas (PwS), which replaces action enumeration with a small set of action templates that the agent instantiates into executable actions. We introduce a cognitive-bandwidth perspective of this trade-off and quantify screening vs. instantiation pressure using context length proxy and failure actions proxy. Across four interactive environments, we identify an inflection point: PwA excels at small interfaces but degrades as action lists grow, whereas PwS becomes increasingly competitive and often superior. A distractor-injection stress test reproduces the PwA drop by enlarging the action space alone, isolating screening load and highlighting schema instantiation as the key remaining bottleneck.</abstract>
<identifier type="citekey">xu-etal-2026-cognitive</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.755/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>17451</start>
<end>17468</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T The Cognitive Bandwidth Bottleneck: Shifting Long-Horizon Agents from Planning with Actions to Planning with Schemas
%A Xu, Baixuan
%A Zheng, Tianshi
%A Wang, Zhaowei
%A Tsang, Hong Ting
%A Wang, Weiqi
%A Fang, Tianqing
%A Song, Yangqiu
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F xu-etal-2026-cognitive
%X Long-horizon LLM agents typically choose from an environment-enumerated list of grounded actions (Planning with Actions, PwA). While this interface reduces formatting risk, scaling the list increases screening load and can bottleneck planning. We study Planning with Schemas (PwS), which replaces action enumeration with a small set of action templates that the agent instantiates into executable actions. We introduce a cognitive-bandwidth perspective of this trade-off and quantify screening vs. instantiation pressure using context length proxy and failure actions proxy. Across four interactive environments, we identify an inflection point: PwA excels at small interfaces but degrades as action lists grow, whereas PwS becomes increasingly competitive and often superior. A distractor-injection stress test reproduces the PwA drop by enlarging the action space alone, isolating screening load and highlighting schema instantiation as the key remaining bottleneck.
%U https://aclanthology.org/2026.emnlp-main.755/
%P 17451-17468
Markdown (Informal)
[The Cognitive Bandwidth Bottleneck: Shifting Long-Horizon Agents from Planning with Actions to Planning with Schemas](https://aclanthology.org/2026.emnlp-main.755/) (Xu et al., EMNLP 2026)
ACL