Skip to content
Preprint

Second Thoughts: Exploring and Planning through Comparative Imagination in World Models

Oct 2026 · 0 citations · 46 references
Computer Science

TL;DR

The introduced SDI (Sibling Disagreement in Imagination), which connects these decisions through a shared actor-based reference to support comparing imagined alternatives against a shared actor reference as a unified approach to exploring and planning with world models.

Abstract

World models let agents compare imagined outcomes before acting, but an agent must still decide which alternatives to explore and when a proposed plan should replace its learned policy. We introduce SDI (Sibling Disagreement in Imagination), which connects these decisions through a shared actor-based reference. SDI compares action sequences imagined from the same model state, using repeated rollouts of the actor's own sequence to measure variation under fixed behavior. This reference scales both the spread across candidate returns, which provides an exploration signal, and a proposed plan's predicted advantage, which governs action acceptance. The agent executes a planned action when its advantage exceeds the reference variability by a fixed factor; otherwise, it follows the actor. SDI reuses the base agent's world model, actor, and critic without requiring an additional disagreement ensemble. Exploration and planning guide training data collection, while evaluation uses the learned actor alone. Across Meta-World, Adroit, and DM-Control, SDI achieves a higher observed task-average score than every compared baseline in each suite and matches or exceeds the strongest baseline's performance on 15 of 18 tasks. Component ablations and robustness analyses further examine the roles of exploration, planning exposure, and action acceptance. Paired evaluations from identical simulator states show that the gate improves both recovery from actor failures and preservation of successful behavior relative to ungated planning, while substantially reducing return degradation. Together, these results support comparing imagined alternatives against a shared actor reference as a unified approach to exploring and planning with world models.

View source

Similar papers

Preprint Nov 2026

Latent Imagination for Parameterized Action Spaces

Parameterized-action MDPs (PAMDPs) ask an agent to choose a discrete action together with continuous parameters whose meaning depends on that choice. Model-free PAMDP methods act fast but need many interactions; model-based methods are sample-efficient but plan at every decision, which makes them slow to act. We introd...

Atahan Cilan, İhsan Mert Muhacıroğlu, Umit Can Bekar · 0 citations
Preprint Nov 2026

Seeing Through the Illusion: Test-Time Scaling for World Action Models

World action models (WAMs) jointly generate actions and imagined futures, but a plausible prediction need not be realized by its co-generated action. We introduce WISE (\emph{World Action Inference Scaling through Evaluation}), a pre-execution selector that scores each sampled rollout on action validity and temporal re...

C. Ahmed, Mahdi Mostajabdaveh, M. Yazdani et al. · 0 citations
Preprint Oct 2026

Train An Imaginative Agent

A seemingly reasonable push in Sokoban can make the puzzle unsolvable. To avoid such mistakes, an agent has to anticipate the consequences of its actions before execution, especially when feedback from the real environment is sparse or actions are irreversible. We introduce Agentic Imagination (Agima), a training frame...

Jia-Xin Ge, Junyi Zhang, Tian-Ning Chai et al. · 0 citations
Preprint Nov 2026

ABHBench: Evaluating Moral Decision-Making of Foundation Models from an Agentic Perspective

Across 12 foundation models, it is found that moral behavior changes substantially under situated agentic context: 75% of models shift significantly between a text-only abstraction and the matched multimodal DBH scenario, and reasoning-enhanced models show no monotonic gains in consistency or human-oriented behavior.

Chao-Ran Li, Yu-Kun Li, Ze-Yuan Zhao et al. · 0 citations
Sep 2026

The Planning Limits of Latent World Models

This work studies action-conditioned predictors built on five frozen self-supervised visual backbones, finding that a world model guides action selection reliably only when the goal lies within, or slightly beyond, the trajectory it imagines during planning.

A. Alrasheed, Basim Azam, Naveed Akhtar · 3 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.