General Agentic Planning Through Simulative Reasoning with World Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Deng, Mingkai, Hou, Jinyu, Hu, Zhiting, Xing, Eric
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911703766138880
author Deng, Mingkai
Hou, Jinyu
Hu, Zhiting
Xing, Eric
author_facet Deng, Mingkai
Hou, Jinyu
Hu, Zhiting
Xing, Eric
contents What does it mean to plan? Current agentic systems, whether scaffolded workflows or end-to-end policies, rely on reactive decision-making: selecting the next action via a fixed procedure with at most undifferentiated adaptive computation (e.g., chain-of-thought) lacking explicit modeling of future outcomes. This limits generalizability, as each new task demands re-engineering rather than transfer of shared reasoning capacity. Humans, by contrast, plan by mentally simulating consequences of candidate actions within an internal world model, a capacity known as simulative reasoning (System II) that supports flexible, goal-directed behavior across diverse contexts. We argue that simulative reasoning through a world model provides a general-purpose planning mechanism for agentic systems, improving upon reactive policies (System I) by grounding decisions in predicted future states rather than pattern-matched responses. To verify this, we introduce SiRA (Simulative Reasoning Architecture), a goal-oriented architecture instantiating simulative reasoning using an LLM-based world model with natural-language belief states, while remaining model-agnostic. We evaluate across three qualitatively distinct task categories: constrained navigation, multi-hop information aggregation, and general instruction following, in a web-browser environment. Across all categories, simulative reasoning achieves up to 124% higher task completion rates than a matched reactive baseline, and increases constrained navigation success from 0% to 32.2% compared to a representative open-web agent. The persistent advantage across distinct task types suggests the benefit stems from generalizable counterfactual evaluation rather than task-specific tuning.
format Preprint
id arxiv_https___arxiv_org_abs_2507_23773
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle General Agentic Planning Through Simulative Reasoning with World Models
Deng, Mingkai
Hou, Jinyu
Hu, Zhiting
Xing, Eric
Artificial Intelligence
Computation and Language
Machine Learning
Robotics
What does it mean to plan? Current agentic systems, whether scaffolded workflows or end-to-end policies, rely on reactive decision-making: selecting the next action via a fixed procedure with at most undifferentiated adaptive computation (e.g., chain-of-thought) lacking explicit modeling of future outcomes. This limits generalizability, as each new task demands re-engineering rather than transfer of shared reasoning capacity. Humans, by contrast, plan by mentally simulating consequences of candidate actions within an internal world model, a capacity known as simulative reasoning (System II) that supports flexible, goal-directed behavior across diverse contexts. We argue that simulative reasoning through a world model provides a general-purpose planning mechanism for agentic systems, improving upon reactive policies (System I) by grounding decisions in predicted future states rather than pattern-matched responses. To verify this, we introduce SiRA (Simulative Reasoning Architecture), a goal-oriented architecture instantiating simulative reasoning using an LLM-based world model with natural-language belief states, while remaining model-agnostic. We evaluate across three qualitatively distinct task categories: constrained navigation, multi-hop information aggregation, and general instruction following, in a web-browser environment. Across all categories, simulative reasoning achieves up to 124% higher task completion rates than a matched reactive baseline, and increases constrained navigation success from 0% to 32.2% compared to a representative open-web agent. The persistent advantage across distinct task types suggests the benefit stems from generalizable counterfactual evaluation rather than task-specific tuning.
title General Agentic Planning Through Simulative Reasoning with World Models
topic Artificial Intelligence
Computation and Language
Machine Learning
Robotics
url https://arxiv.org/abs/2507.23773