🛰️ Daily AI Frontier
‹ back to 2026-08-23

Orthogonal JEPA: Factorized Predictive States for Latent World Models

Research LLMs & Foundation Models

Ranking

Overall 79
Content 95
Popularity 41

Observed public metrics from 1 member.

Merged summary

TL;DR - Orthogonal JEPA is a latent world-modeling framework that factorizes predictive states into multiple orthogonal components rather than using one monolithic target embedding. The design aims to reduce redundant capacity, preserve less-dominant predictive structure, and support more stable forecasting, planning, and rollouts.

  • Learned basis matrices decompose each target state, while dedicated branches predict each component from shared context.
  • Predictive regression preserves factor magnitudes so the components can be synthesized back into a complete latent state.
  • Orthogonality, factor-activity, and online variance objectives discourage repeated directions, inactive factors, and encoder collapse.
  • The framework covers temporal, spatial, and partial-observation prediction and is evaluated across vision, biology, healthcare, control, and molecular dynamics.

Sources (1)

Orthogonal JEPA: Factorized Predictive States for Latent World Models

arXiv cs.LG Taoyong Cui, Pheng Ann Heng, Wanli Ouyang 2026-08-20 arXiv:2608.20065
Public signals Semantic Scholar citations 0 · Semantic Scholar influential citations 0
Providers: Hugging Face · N/A OpenAlex · N/A Publisher · N/A Semantic Scholar · Citations 0 · Influential citations 0 X · N/A Fetched 2026-08-25 14:14:00.002375 UTC

TL;DR - Orthogonal JEPA is a latent world-modeling framework that factorizes predictive states into multiple orthogonal components rather than using one monolithic target embedding. The design aims to reduce redundant capacity, preserve less-dominant predictive structure, and support more stable forecasting, planning, and rollouts.

  • Learned basis matrices decompose each target state, while dedicated branches predict each component from shared context.
  • Predictive regression preserves factor magnitudes so the components can be synthesized back into a complete latent state.
  • Orthogonality, factor-activity, and online variance objectives discourage repeated directions, inactive factors, and encoder collapse.
  • The framework covers temporal, spatial, and partial-observation prediction and is evaluated across vision, biology, healthcare, control, and molecular dynamics.
item →