{"ID":2844720,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.06136","arxiv_id":"2511.06136","title":"When Object-Centric World Models Meet Policy Learning: From Pixels to Policies, and Where It Breaks","abstract":"Object-centric world models (OCWM) aim to decompose visual scenes into object-level representations, providing structured abstractions that could improve compositional generalization and data efficiency in reinforcement learning. We hypothesize that explicitly disentangled object-level representations, by localizing task-relevant information, can enhance policy performance across novel feature combinations. To test this hypothesis, we introduce DLPWM, a fully unsupervised, disentangled object-centric world model that learns object-level latents directly from pixels. DLPWM achieves strong reconstruction and prediction performance, including robustness to several out-of-distribution (OOD) visual variations. However, when used for downstream model-based control, policies trained on DLPWM latents underperform compared to DreamerV3. Through latent-trajectory analyses, we identify representation shift during multi-object interactions as a key driver of unstable policy learning. Our results suggest that, although object-centric perception supports robust visual modeling, achieving stable control requires mitigating latent drift.","short_abstract":"Object-centric world models (OCWM) aim to decompose visual scenes into object-level representations, providing structured abstractions that could improve compositional generalization and data efficiency in reinforcement learning. We hypothesize that explicitly disentangled object-level representations, by localizing ta...","url_abs":"https://arxiv.org/abs/2511.06136","url_pdf":"https://arxiv.org/pdf/2511.06136v2","authors":"[\"Stefano Ferraro\",\"Akihiro Nakano\",\"Masahiro Suzuki\",\"Yutaka Matsuo\"]","published":"2025-11-08T21:09:44Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Reinforcement Learning\"]","has_code":false}