Metraipolis › archive › dc_d7b81e291bc2

CAER: Causal Action Effect Reweighting for World Model Training — arXiv:2608.30897

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.30897
License (as archived)
CC0-1.0
Archived
2026-10-17T07:44:58.021Z
SHA-256 of the archived text
31f119f0954a0974d1811decdfebcb599a9a2c9c61a156eec9a9963caf5b7675
Stable URL of this page
https://metraipolis.com/documento/dc_d7b81e291bc2

How to cite

CAER: Causal Action Effect Reweighting for World Model Training — arXiv:2608.30897 — arXiv (en), https://arxiv.org/abs/2608.30897 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d7b81e291bc2 (Metraipolis archive dc_d7b81e291bc2, sha256 31f119f0954a0974d1811decdfebcb599a9a2c9c61a156eec9a9963caf5b7675)
BibTeX
@misc{dcd7b81e291bc2,
  title = {CAER: Causal Action Effect Reweighting for World Model Training — arXiv:2608.30897},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.30897},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d7b81e291bc2; sha256 31f119f0954a0974d1811decdfebcb599a9a2c9c61a156eec9a9963caf5b7675},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.30897. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

World models are becoming core infrastructure for embodied intelligence, with action-conditioned video generation providing controllable predictions of how scenes evolve after agent interventions. Yet existing models are commonly trained with space-time-uniform mean squared error, allowing abundant background tokens to dominate the gradient while sparse interaction dynamics remain under-optimized; such uniform fitting rewards reconstructing appearance rather than learning how actions change the world. We introduce Causal Action Effect Reweighting (CAER), a general training paradigm that redistributes supervision toward the tokens whose predicted future is causally affected by the action. CAER contrasts the model's own predictions with and without action conditioning to localize these tokens online, then normalizes the resulting effect map into a weight that preserves the total coefficient mass and changes only where it is spent. This online signal requires no external annotations or offline preprocessing, avoids additional data-processing time, and scales naturally with model and dataset size. Experiments across heterogeneous action-conditioned world-model tasks show that CAER converges to better solutions than uniform MSE training, with consistent improvements in the physical consistency, controllability, and visual quality of generated videos.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.30897 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T07:28:26.630Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).