# Two New Preprints Push World Models Toward Physical Quantities Rather Than Pixel Reconstruction

One proposes spectral targets to structure the latent space of JEPA-style predictors, the other releases a video dataset of falling coffee grounds paired with per-frame scale readings as ground truth for weight estimation.

- Published: 2026-09-07T06:22:00.097Z
- Canonical: https://polylog.news/ai/2026-09-07/two-new-preprints-push-world-models-toward-physical-quantiti
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [Polylog editors](https://polylog.news)

Latent world models predict and plan using a compressed representation rather than raw pixels, which is what makes them computationally practical for control tasks. The open problem is that nothing in the training objective forces that late…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-09-07/two-new-preprints-push-world-models-toward-physical-quantiti (subscription information: https://polylog.news/pricing).