@inproceedings{yang-etal-2025-efficient,
title = "Efficient Integration of External Knowledge to {LLM}-based World Models via Retrieval-Augmented Generation and Reinforcement Learning",
author = "Yang, Chang and
Wang, Xinrun and
Zhang, Qinggang and
Jiang, Qi and
Huang, Xiao",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-emnlp.504/",
doi = "10.18653/v1/2025.findings-emnlp.504",
pages = "9484--9501",
ISBN = "979-8-89176-335-7",
abstract = "World models achieve remarkable success in predicting future states and planning in complex environments and Large Language Models (LLMs) serve as promising foundation to build general world models. However, their performances are usually constrained by the limited external knowledge to specific environments. Existing research attempts to enhance LLM-based world models through prompting or fine-tuning approaches, which are either requiring human knowledge or computationally extensive. Therefore, we introduce Retrieval-Augmented World Models (RAWM), a novel framework that leverages retrieval-augmented generation to efficiently integrate the external knowledge to LLM-based world models. Our main contributions are threefold: (i) We introduce a memory system and design an embedding model to retrieve relevant experiences as the in-context examples to improve the world model{'}s predictive accuracy. (ii) We develop a reinforcement learning (RL) training pipeline that fine-tunes a small MLP head on the pre-trained embedding model using Proximal Policy Optimization (PPO), further enhancing prediction performance. (iii) We conduct extensive experiments across three diverse environments, i.e., Game24, BlocksWorld, and BabyAI, demonstrating that RAWM consistently outperforms baseline models and exhibits strong generalizability. By leveraging the retrieval-augmented generation and the efficient RL training pipeline, RAWM dynamically utilizes relevant historical experiences and equips LLMs with environment-specific external knowledge without retraining, enabling more accurate and generalizable predictions."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yang-etal-2025-efficient">
<titleInfo>
<title>Efficient Integration of External Knowledge to LLM-based World Models via Retrieval-Augmented Generation and Reinforcement Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chang</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xinrun</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qinggang</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qi</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiao</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-335-7</identifier>
</relatedItem>
<abstract>World models achieve remarkable success in predicting future states and planning in complex environments and Large Language Models (LLMs) serve as promising foundation to build general world models. However, their performances are usually constrained by the limited external knowledge to specific environments. Existing research attempts to enhance LLM-based world models through prompting or fine-tuning approaches, which are either requiring human knowledge or computationally extensive. Therefore, we introduce Retrieval-Augmented World Models (RAWM), a novel framework that leverages retrieval-augmented generation to efficiently integrate the external knowledge to LLM-based world models. Our main contributions are threefold: (i) We introduce a memory system and design an embedding model to retrieve relevant experiences as the in-context examples to improve the world model’s predictive accuracy. (ii) We develop a reinforcement learning (RL) training pipeline that fine-tunes a small MLP head on the pre-trained embedding model using Proximal Policy Optimization (PPO), further enhancing prediction performance. (iii) We conduct extensive experiments across three diverse environments, i.e., Game24, BlocksWorld, and BabyAI, demonstrating that RAWM consistently outperforms baseline models and exhibits strong generalizability. By leveraging the retrieval-augmented generation and the efficient RL training pipeline, RAWM dynamically utilizes relevant historical experiences and equips LLMs with environment-specific external knowledge without retraining, enabling more accurate and generalizable predictions.</abstract>
<identifier type="citekey">yang-etal-2025-efficient</identifier>
<identifier type="doi">10.18653/v1/2025.findings-emnlp.504</identifier>
<location>
<url>https://aclanthology.org/2025.findings-emnlp.504/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>9484</start>
<end>9501</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Efficient Integration of External Knowledge to LLM-based World Models via Retrieval-Augmented Generation and Reinforcement Learning
%A Yang, Chang
%A Wang, Xinrun
%A Zhang, Qinggang
%A Jiang, Qi
%A Huang, Xiao
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Findings of the Association for Computational Linguistics: EMNLP 2025
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-335-7
%F yang-etal-2025-efficient
%X World models achieve remarkable success in predicting future states and planning in complex environments and Large Language Models (LLMs) serve as promising foundation to build general world models. However, their performances are usually constrained by the limited external knowledge to specific environments. Existing research attempts to enhance LLM-based world models through prompting or fine-tuning approaches, which are either requiring human knowledge or computationally extensive. Therefore, we introduce Retrieval-Augmented World Models (RAWM), a novel framework that leverages retrieval-augmented generation to efficiently integrate the external knowledge to LLM-based world models. Our main contributions are threefold: (i) We introduce a memory system and design an embedding model to retrieve relevant experiences as the in-context examples to improve the world model’s predictive accuracy. (ii) We develop a reinforcement learning (RL) training pipeline that fine-tunes a small MLP head on the pre-trained embedding model using Proximal Policy Optimization (PPO), further enhancing prediction performance. (iii) We conduct extensive experiments across three diverse environments, i.e., Game24, BlocksWorld, and BabyAI, demonstrating that RAWM consistently outperforms baseline models and exhibits strong generalizability. By leveraging the retrieval-augmented generation and the efficient RL training pipeline, RAWM dynamically utilizes relevant historical experiences and equips LLMs with environment-specific external knowledge without retraining, enabling more accurate and generalizable predictions.
%R 10.18653/v1/2025.findings-emnlp.504
%U https://aclanthology.org/2025.findings-emnlp.504/
%U https://doi.org/10.18653/v1/2025.findings-emnlp.504
%P 9484-9501
Markdown (Informal)
[Efficient Integration of External Knowledge to LLM-based World Models via Retrieval-Augmented Generation and Reinforcement Learning](https://aclanthology.org/2025.findings-emnlp.504/) (Yang et al., Findings 2025)
ACL