AI Paper Atlas

Research map / Multi-agent systems and games / Human-AI and goal-conditioned

Learning Task-Sufficient World Models by Synergizing Agentic Exploration and Structured Modeling

Fan Feng, Yujia Zheng, Minghao Fu, Yongqiang Chen, Guangyi Chen, Kevin Patrick Murphy, Biwei Huang, Kun Zhang

ICML 2026 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: World-model-based agents · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use MIST-WM latent masking with agentic probing instead of random data collection and unstructured latents (ablation)MIST-WM · RoboSuite Stacking probingaverage R2: 0.31 -> 0.78Improved
collect data with METRA skills instead of random (ablation)world model data curation · RoboSuite Stacking probingaverage R2: 0.31 -> 0.59Improved
use structured representation learning (I-Factor, Denoised MDP) instead of DINO-WM (ablation)world model backbone · RoboSuite Stacking probingaverage R2: 0.40 -> 0.48 / 0.56Improved
add MI term in the minimality score (ablation)MIST-WM · single-task and generalization (normalized to Oracle)normalized scoreImproved
add masking in the minimality score (ablation)MIST-WM · generalization tasksnormalized scoreImproved
add sufficiency term (ablation)MIST-WM · single-task and generalizationnormalized scoreImproved
add adaptive curriculum instead of random task order (ablation)MIST-WM · generalization vs single-tasknormalized scoreImproved
replace Dreamer-v3 encoder with DINO-WM or R3M features (ablation)MIST-WM · ablation tasksnormalized scoreMixed
use MIST-WM instead of TD-MPC2DMControl single-task · Reacher-Hardepisodic rewardHurt
add active probing skills with random-vs-curiosity data (ablation)MIST-WM exploration · generalization tasksnormalized scoreMixed
add MIST-WM structured modeling on top of Dreamer-V3Dreamer-V3 · training costGPU days: 0.25 -> 0.35 (DMControl)Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy