Paper Detail

Decision-Metric Alignment in Latent World Models: Diagnostics and Action-Conditioned Objectives for MPC Planning

Jiawei Wang, Ke Rui, Yushen Zuo, Yichun Feng, Minglei Li

huggingface Score 4.0

Published 2026-08-19 · First seen 2026-08-20

General AI

Abstract

JEPA-style latent world models can use Euclidean distance to a goal latent as the cost for model-predictive control (MPC). Strong decoding of task variables, however, does not guarantee that this particular cost ranks candidate action sequences by real task progress. We call the latter property decision-metric alignment. We introduce Plan-Real Spearman, which measures latent--real rank agreement on random plans, and CEM-stage Spearman, which measures the same agreement as cross-entropy-method (CEM) search concentrates its proposal. We analyze sufficient conditions under which latent distance preserves real-cost rankings, identifying encoder distortion, terminal rollout error, and candidate margins as the controlling quantities. Guided by the observed empirical alignment gap, DA-LeWM augments LeWM with inverse-dynamics and demonstration-conditioned goal-action heads. Across all our experiments, DA-LeWM accelerates convergence and achieves higher online success than LeWM, while probe scores remain similar. These results show that action-conditioned objectives improve the geometry used by Euclidean-cost, CEM-based latent MPC.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
later
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@misc{wang2026decision,
  title = {Decision-Metric Alignment in Latent World Models: Diagnostics and Action-Conditioned Objectives for MPC Planning},
  author = {Jiawei Wang and Ke Rui and Yushen Zuo and Yichun Feng and Minglei Li},
  year = {2026},
  abstract = {JEPA-style latent world models can use Euclidean distance to a goal latent as the cost for model-predictive control (MPC). Strong decoding of task variables, however, does not guarantee that this particular cost ranks candidate action sequences by real task progress. We call the latter property decision-metric alignment. We introduce Plan-Real Spearman, which measures latent--real rank agreement on random plans, and CEM-stage Spearman, which measures the same agreement as cross-entropy-method (C},
  url = {https://huggingface.co/papers/2608.18746},
  keywords = {JEPA, latent world models, model-predictive control, decision-metric alignment, Plan-Real Spearman, CEM-stage Spearman, encoder distortion, DA-LeWM, inverse-dynamics, demonstration-conditioned goal-action, huggingface daily},
  eprint = {2608.18746},
  archiveprefix = {arXiv},
}

Metadata

{}