@inproceedings{ranaldi-2026-world,
title = "World-Model Evolving Agents",
author = "Ranaldi, Leonardo",
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.2167/",
pages = "49279--49294",
ISBN = "979-8-89176-459-0",
abstract = "Agents learn from experience, distilling it into memories, skills, or abstractions that improve which action they propose. The consequence of that action is discovered only after execution, when a long-horizon error has already become irreversible. We introduce WEVA (World-model Evolving Agents), a framework that makes gained experience predictive. WEVA learns from interaction traces what information is predictive and how actions transform the resulting state, using a quasi-symbolic representation over entities, preconditions, and outcomes. Specifically, before acting, EVA predicts the states induced by candidate action sequences, discards those that violate learned preconditions, and executes the first action of the highest-scoring surviving branch. The agreement between predicted and realised states delivers a model-trust signal that regulates when lookahead is reliable and when further interaction is more informative. WEVA improves task success while executing fewer actions, reduces trajectories that terminate in externally verified constraint violations, and detects changes in environmental dynamics before they are reflected in task success. These results position WEVA as a step towards agents that predict the consequences of their decisions and verify them against what they hold about the world before acting."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ranaldi-2026-world">
<titleInfo>
<title>World-Model Evolving Agents</title>
</titleInfo>
<name type="personal">
<namePart type="given">Leonardo</namePart>
<namePart type="family">Ranaldi</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>Agents learn from experience, distilling it into memories, skills, or abstractions that improve which action they propose. The consequence of that action is discovered only after execution, when a long-horizon error has already become irreversible. We introduce WEVA (World-model Evolving Agents), a framework that makes gained experience predictive. WEVA learns from interaction traces what information is predictive and how actions transform the resulting state, using a quasi-symbolic representation over entities, preconditions, and outcomes. Specifically, before acting, EVA predicts the states induced by candidate action sequences, discards those that violate learned preconditions, and executes the first action of the highest-scoring surviving branch. The agreement between predicted and realised states delivers a model-trust signal that regulates when lookahead is reliable and when further interaction is more informative. WEVA improves task success while executing fewer actions, reduces trajectories that terminate in externally verified constraint violations, and detects changes in environmental dynamics before they are reflected in task success. These results position WEVA as a step towards agents that predict the consequences of their decisions and verify them against what they hold about the world before acting.</abstract>
<identifier type="citekey">ranaldi-2026-world</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2167/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>49279</start>
<end>49294</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T World-Model Evolving Agents
%A Ranaldi, Leonardo
%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 ranaldi-2026-world
%X Agents learn from experience, distilling it into memories, skills, or abstractions that improve which action they propose. The consequence of that action is discovered only after execution, when a long-horizon error has already become irreversible. We introduce WEVA (World-model Evolving Agents), a framework that makes gained experience predictive. WEVA learns from interaction traces what information is predictive and how actions transform the resulting state, using a quasi-symbolic representation over entities, preconditions, and outcomes. Specifically, before acting, EVA predicts the states induced by candidate action sequences, discards those that violate learned preconditions, and executes the first action of the highest-scoring surviving branch. The agreement between predicted and realised states delivers a model-trust signal that regulates when lookahead is reliable and when further interaction is more informative. WEVA improves task success while executing fewer actions, reduces trajectories that terminate in externally verified constraint violations, and detects changes in environmental dynamics before they are reflected in task success. These results position WEVA as a step towards agents that predict the consequences of their decisions and verify them against what they hold about the world before acting.
%U https://aclanthology.org/2026.emnlp-main.2167/
%P 49279-49294
Markdown (Informal)
[World-Model Evolving Agents](https://aclanthology.org/2026.emnlp-main.2167/) (Ranaldi, EMNLP 2026)
ACL
- Leonardo Ranaldi. 2026. World-Model Evolving Agents. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 49279–49294, Budapest, Hungary. Association for Computational Linguistics.