Explorative Modeling: Unlocking A New Scaling Axis and End-to-End Generative Modeling

1UIUC   2Harvard

TL;DR

Explorative Modeling (XM) is a new paradigm for handling multimodality in generative models: instead of factoring the generation procedure into smaller steps during training (the trick every existing scalable approach relies on), a model explores many candidate generations and reinforces only the best-matching one, so its predictions capture individual modes instead of blurring them. This is useful in two distinct ways. First, exploration is a new scaling axis for existing generative models, yielding up to 53% better FLOP efficiency, 2.5× better data efficiency, and 47% better parameter efficiency—and, like data and parameters, these gains grow monotonically with scale. Second, as a standalone model, XM is the first reconstructive generative model to succeed with end-to-end training, achieving 20–250× inference speedups while matching diffusion on certain tasks.

Figures & Visualizations

Under construction — visualizations of the Explorative Modeling process will be added here soon.

Abstract

The paper, code, and additional resources are under construction and will be released soon. Check back for updates!

BibTeX

@misc{gladstone2026explorativemodeling,
  title={Explorative Modeling: Unlocking A New Scaling Axis and End-to-End Generative Modeling},
  author={Alexi Gladstone and Heng Ji and Yilun Du},
  year={2026},
  note={Preprint. Under review.}
}