AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Object-Centric World Models for Causality-Aware Reinforcement Learning

Yosuke Nishimoto, Takashi Matsubara

AAAI 2026 · Reinforcement learning › Visual and representation RL · approach: Latent-space world models · 3 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
add object-centric world model alone (no object-centric policy/value) (ablation)STICA · Safety Gym PointButton1, CarButton1returnNo clear effect
add Transformer policy and value networks over object tokens (ablation)STICA · Safety Gym PointButton1, CarButton1returnImproved
add background removal via static background latent (ablation)STICA · Safety Gym PointButton1, CarButton1returnImproved
add causal attention in policy and value networks (ablation)STICA · Safety Gym PointButton1, CarButton1returnImproved
use STICA instead of DreamerV3STICA · Safety Gym PointButton1return: 4.01 -> 11.52Improved
use STICA instead of DreamerV3STICA · Safety Gym PointGoal1return: 19.13 -> 13.63Hurt
use STICA instead of TWMSTICA · OCVRL Object Interactionsuccess rate: 0.080 -> 0.333Improved
use STICA instead of best baselineSTICA · Safety Gym (normalized mean)normalized mean return: 4.06 -> 5.49Improved
use STICA instead of DreamerV3STICA · Safety Gym PointGoal2return: 13.64 -> 13.64No clear effect

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