REVIEW 4 cited by
Reduce, Reuse, Recycle: Compositional Generation with Energy-Based Diffusion Models and MCMC
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Since their introduction, diffusion models have quickly become the prevailing approach to generative modeling in many domains. They can be interpreted as learning the gradients of a time-varying sequence of log-probability density functions. This interpretation has motivated classifier-based and classifier-free guidance as methods for post-hoc control of diffusion models. In this work, we build upon these ideas using the score-based interpretation of diffusion models, and explore alternative ways to condition, modify, and reuse diffusion models for tasks involving compositional generation and guidance. In particular, we investigate why certain types of composition fail using current techniques and present a number of solutions. We conclude that the sampler (not the model) is responsible for this failure and propose new samplers, inspired by MCMC, which enable successful compositional generation. Further, we propose an energy-based parameterization of diffusion models which enables the use of new compositional operators and more sophisticated, Metropolis-corrected samplers. Intriguingly we find these samplers lead to notable improvements in compositional generation across a wide set of problems such as classifier-guided ImageNet modeling and compositional text-to-image generation.
Forward citations
Cited by 4 Pith papers
-
Text Dictates, Music Decorates: Energy-based Attention for Editable Dance Motion Generation
STREAM decouples text (via AdaLN) from music (via energy-based BEAM attention) to generate editable, musically aligned dance motions with a new annotated dataset and editability metric.
-
Compositional Scene Understanding through Inverse Generative Modeling
Composing per-concept diffusion models and inverting them with denoising loss enables multi-object scene understanding that generalizes beyond the training distribution.
-
TILT: Improving Compositional Generation in Diffusion Models with a Model-Intrinsic Reward
A training-free, test-time guidance rule that tilts a diffusion model's samples toward regions where every concept in a prompt is jointly present; it improves several T2ICompBench categories over prior correctors and ...
-
Combining complex Langevin dynamics with score-based and energy-based diffusion models
Energy-based diffusion models trained on complex Langevin data produce an explicit energy function for the sampled distribution, enabling MCMC without re-simulation.
Discussion (0). Continue with ORCID to comment.