Unconditional Score Estimation For Guidance

via “classifier-free guidance for prompt adherence control”

text-to-image model by undefined. 6,21,488 downloads.

Unique: Implements guidance as a post-hoc scaling of noise predictions rather than modifying the model architecture, enabling zero-shot control without retraining. Guidance scale is a continuous hyperparameter, allowing fine-grained tradeoffs between prompt adherence and diversity.

vs others: More flexible and computationally efficient than explicit classifier-based guidance (which requires a separate classifier model); provides intuitive control compared to prompt engineering alone.

via “classifier-free guidance with dynamic guidance scaling”

Official repository for LTX-Video

Unique: Implements dynamic per-timestep guidance scaling with optional schedule control, enabling fine-grained trade-offs between prompt adherence and output quality, vs. static guidance scales used in most competing approaches

vs others: Dynamic guidance scheduling provides better quality than static guidance by using strong guidance early (for structure) and weak guidance late (for detail), improving visual quality by ~15-20% vs. constant guidance scales

* ⭐ 08/2022: [Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation (DreamBooth)](https://arxiv.org/abs/2208.12242)

Unique: Learns unconditional scores through conditioning dropout (masking signals during training) rather than training separate models, enabling efficient joint learning within a single parameter space

vs others: More parameter-efficient than separate unconditional models, but may produce biased unconditional scores compared to dedicated unconditional models trained on pure unconditional data

via “classifier-free guidance with dynamic weighting”

IF — AI demo on HuggingFace

Unique: Uses classifier-free guidance (training on both conditioned and unconditional samples) rather than requiring a separate classifier or reward model, enabling efficient guidance without additional model components.

vs others: Simpler to implement and train than classifier-based guidance (no separate classifier needed) while providing more flexible control than fixed-weight conditioning.