REVIEW 7 cited by
Dirichlet Flow Matching with Applications to DNA Sequence Design
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
Signed reviews
abstract
Discrete diffusion or flow models could enable faster and more controllable sequence generation than autoregressive models. We show that na\"ive linear flow matching on the simplex is insufficient toward this goal since it suffers from discontinuities in the training target and further pathologies. To overcome this, we develop Dirichlet flow matching on the simplex based on mixtures of Dirichlet distributions as probability paths. In this framework, we derive a connection between the mixtures' scores and the flow's vector field that allows for classifier and classifier-free guidance. Further, we provide distilled Dirichlet flow matching, which enables one-step sequence generation with minimal performance hits, resulting in $O(L)$ speedups compared to autoregressive models. On complex DNA sequence generation tasks, we demonstrate superior performance compared to all baselines in distributional metrics and in achieving desired design targets for generated sequences. Finally, we show that our classifier-free guidance approach improves unconditional generation and is effective for generating DNA that satisfies design targets. Code is available at https://github.com/HannesStark/dirichlet-flow-matching.
Forward citations
Cited by 7 Pith papers
-
What Exactly Does Guidance Do in Masked Discrete Diffusion Models
With exact scores and no discretization error, CFG in 1D masked discrete diffusion samples exactly the tilted distribution; in 2D it does not, and the TV convergence rate is double-exponential in guidance strength.
-
Should the Boundary Term Be Learned in Reflected Diffusion? Conormal Trace and Reflection Masking
For reflected diffusion on bounded domains, the no-flux condition fixes one scalar per boundary point, the conormal trace of the score, and the paper gives an exact box parametrization, proves a misspecification floor...
-
Stiefel Flow Matching for Moment-Constrained Structure Elucidation
A generative model on the Stiefel manifold predicts molecular 3D structures that exactly match given moments of inertia, improving success rates and sampling cost over Euclidean diffusion baselines.
-
Concurrent Image Understanding and Generation: Self-Correcting Coupled Markov Jump Processes
CO2Jump couples text and image denoising through cross-modal attention and remasking, achieving best joint accuracy on three concurrent-generation tasks.
-
Exploring Discrete Flow Matching for 3D De Novo Molecule Generation
FlowMol-CTMC, using discrete-state continuous-time Markov chain flows, generates 3D molecules with higher stability and validity than prior 3D generation models while using far fewer parameters.
-
Simple Guidance Mechanisms for Discrete Diffusion Models
Uniform-noise discrete diffusion trained with a continuous-time variational bound (UDLM) plus discrete classifier-free and classifier-based guidance improves controllable generation over autoregressive baselines on ge...
-
MolPIF: A Parameter Interpolation Flow Model for Molecule Generation
MolPIF generates 3D ligands by interpolating the parameters of Gaussian coordinate and Dirichlet atom-type distributions, reporting stronger docking scores and geometric fidelity than prior flow and diffusion models o...
Discussion (0). Continue with ORCID to comment.