In-situ Autoguidance: Eliciting Self-Correction in Diffusion Models
The generation of high-quality, diverse, and prompt-aligned images is a central goal in image-generating diffusion models. The popular classifier-free guidance (CFG) approach improves quality and alignment at the cost of reduced variation, creating an inherent entanglement of these effects. Recent work has successfully disentangled these properties by guiding a model with a separately trained, inferior counterpart; however, this solution introduces the considerable overhead of requiring an auxiliary model. We challenge this prerequisite by introducing In-situ Autoguidance, a method that elicits guidance from the model itself without any auxiliary components. Our approach dynamically generates an inferior prediction on the fly using a stochastic forward pass, reframing guidance as a form of inference-time self-correction. We demonstrate that this zero-cost approach is not only viable but also establishes a powerful new baseline for cost-efficient guidance, proving that the benefits of self-guidance can be achieved without external models.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Autoguided Online Data Curation for Diffusion Model Training
The costs of generative model compute rekindled promises and hopes for efficient data curation. In this work, we investigate whether recently developed autoguidance and online data selection methods can improve the time …
Synthetic Data GenerationImage GenerationSynthetic Error Injection Fails to Elicit Self-Correction In Language Models
Reinforcement learning has become the dominant paradigm for eliciting reasoning and self-correction capabilities in large language models, but its computational expense motivates exploration of alternatives. Inspired by …
Reinforcement LearningAutonomous DrivingSpatiotemporal Skip Guidance for Enhanced Video Diffusion Sampling
Diffusion models have emerged as a powerful tool for generating high-quality images, videos, and 3D content. While sampling guidance techniques like CFG improve quality, they reduce diversity and motion. Autoguidance mit…
DiversityDo LLMs Need Inherent Reasoning Before Reinforcement Learning? A Study in Korean Self-Correction
Large Language Models (LLMs) demonstrate strong reasoning and self-correction abilities in high-resource languages like English, but their performance remains limited in low-resource languages such as Korean. In this stu…
Reinforcement LearningMathematical ReasoningFine-Tuning Masked Diffusion for Provable Self-Correction
A natural desideratum for generative models is self-correction--detecting and revising low-quality tokens at inference. While Masked Diffusion Models (MDMs) have emerged as a promising approach for generative modeling in…