This repository is a curated collection of research papers and resources focused on controllable generation with text-to-image diffusion models. It serves as a comprehensive survey and reference for researchers and practitioners in the field of generative AI, aiming to organize and categorize advancements in novel conditional generation techniques.

How It Works

The project acts as a living bibliography, meticulously cataloging papers that explore various methods for controlling text-to-image diffusion models. It categorizes these methods by the type of control introduced, such as personalization, style, interaction, image-driven, distribution-driven, and spatial control, providing a structured overview of the research landscape.

Quick Start & Requirements

This repository is a curated list of papers and does not have a direct installation or execution command. It requires no specific software to "run" but serves as a knowledge base.

Highlighted Details

• Comprehensive categorization of controllable generation techniques, including personalization, subject-driven, style-driven, interaction-driven, image-driven, distribution-driven, and spatial control.
• Extensive list of recent research papers (primarily from 2023-2024) with direct PDF links, covering a wide array of control mechanisms and applications.
• Includes sections on advanced text-conditioned generation, in-context generation, brain-guided generation, sound-guided generation, and text rendering.
• Provides a clear citation for the associated survey paper, facilitating academic referencing.

Maintenance & Community

The repository is maintained by PRIV-Creation and encourages contributions via GitHub issues to add new papers, rather than direct pull requests.

Licensing & Compatibility

The repository itself does not specify a license, but the linked research papers are subject to their respective publication licenses.

Limitations & Caveats

As a curated list, the repository does not provide code or implementations for the discussed techniques. Its value is purely informational, requiring users to seek out individual research papers for practical application.