Grounded Iterative Language Planning: How Parameterized World Models Reduce Hallucination Propagation in LLM Agents
This paper compares agent-based and parameterized world models for LLM agents, proposing GILP, which combines a small parameterized backbone with LLM reasoning to reduce hallucinated state rates from 0.176 to 0.035 and increase success rates from 0.668 to 0.838 with only ~22% extra LLM calls.
[2606.27806] Grounded Iterative Language Planning: How Parameterized World Models Reduce Hallucination Propagation in LLM Agents
[Submitted on 26 Jun 2026]
Title:Grounded Iterative Language Planning: How Parameterized World Models Reduce Hallucination Propagation in LLM Agents
View a PDF of the paper titled Grounded Iterative Language Planning: How Parameterized World Models Reduce Hallucination Propagation in LLM Agents, by Xinyuan Song and 1 other authors
View PDF HTML (experimental)
Abstract:World models for language agents come in two useful forms. An agent-based world model calls an LLM API and reasons flexibly in language, but its errors appear as hallucinated state changes that are hard to score with ordinary regression losses. A parameterized world model is a trained transition predictor; its errors are easier to measure with quantities such as NodeMSE, delta accuracy, and validity accuracy, but it is usually weaker as a standalone planner. We compare these two families on four graph-structured planning benchmarks and introduce operational hallucination metrics for the agent-based case. The comparison motivates \textbf{Grounded Iterative Language Planning} (GILP), which trains only a small parameterized backbone and combines it with API-based agent reasoning. The backbone supplies valid actions, predicted state deltas, risk, and value; the LLM drafts an action and imagined delta; and a consistency gate asks for revision when the two disagree. On real GPT-4o-mini calls, GILP reduces hallucinated-state rate from 0.176 to 0.035. In calibrated simulator ablations, it raises success from 0.668 to 0.838 while adding only ~22% extra LLM calls.
Comments: Under Review
Subjects:
Artificial Intelligence (cs.AI)
Cite as: arXiv:2606.27806 [cs.AI]
(or arXiv:2606.27806v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2606.27806
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Zekun Cai [view email] [v1] Fri, 26 Jun 2026 07:45:15 UTC (1,414 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Grounded Iterative Language Planning: How Parameterized World Models Reduce Hallucination Propagation in LLM Agents, by Xinyuan Song and 1 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.AI
new | recent | 2026-06
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
Loading...
Data provided by:
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)