cs.LGNov 4, 2026

MAC-v2: Long-horizon model-based value expansion from pixels

Authors: Kwanyoung Park, Sergey Levine

Organizations: UC Berkeley

Abstract

We study whether long-horizon model-based value expansion can serve as an effective framework in pixel-based offline reinforcement learning (RL). Long-horizon model-based value expansion from pixels is substantially harder than state-based environments, as the agent must learn both the representation and its dynamics, and errors from either can accumulate recursively over imagined trajectories. We introduce MAC-v2, which enables long-horizon model-based expansion from pixels by training a reconstruction-free latent world model with action chunking. By learning representations from prediction and control objectives, MAC-v2 does not need to encode control-irrelevant visual details that can hinder the dynamics prediction. Action chunking further mitigates the prediction error by reducing the number of recursive model predictions required for value expansion. Across diverse visual benchmarks including visual OGBench, MMBench2, Robomimic, and LIBERO-10, MAC-v2 achieves strong performance against model-free and model-based offline RL baselines. Our analyses show that long-horizon model-based value expansion is critical for the performance of MAC-v2, and that both action chunking and latent learning objectives are essential for reliable long-horizon latent imagination.

Figures & tables

Appendix figures & tables15 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. ELVIS: Ensemble-Calibrated Latent Imagination for Long-Horizon Visual MPC

    May 6, 2026Yurui Du, Pinhao Song, Yutong Hu +1Visual SSMsModel Predictive Control

  2. Beyond a single latent space: a dual-latent world model for long-horizon planning

    Sep 29, 2026Delin Zhao, Zhengrong Yue, Shaobin Zhuang +6World Model-Based PlanningGoal-Conditioned Planning

  3. VPFlow: Value-Predictive Planning with Imperfect World Models for Offline Goal Reaching

    Oct 19, 2026Yunzhe Hu, Dong XuOffline RLModel-Based Planning