SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2502.01639 · ICCV · 2025

SliderSpace: Decomposing the Visual Capabilities of Diffusion Models

Richard Zhang, Eli Shechtman, David Bau, Rohit Gandikota, Zongze Wu, Nick Kolkin

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Figure 1. Given a prompt, SliderSpace identifies the principal directions of the visual capabilities of a diffusion model by decomposing the image distribution over the prompt. By manipulating these directions as sliders, users can control and combine them to explore the creative possibilities of the model. We visualize the top directions discovered by SliderSpace within Flux Schnell [4] for the concept "Toy".

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2502.01639")
get_code_for_paper("2502.01639")
have("2502.01639")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2502.01639.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2502.01639.svg)](https://syntology.ai/paper/2502.01639)

Connect an agent — have() is free.