Pith. sign in

REVIEW 17 cited by

Generative Flows on Discrete State-Spaces: Enabling Multimodal Flows with Applications to Protein Co-Design

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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

arxiv 2402.04997 v2 pith:3HKYS33U submitted 2024-02-07 stat.ML cs.LGq-bio.QM

Andrew Campbell , Jason Yim , Regina Barzilay , Tom Rainforth , Tommi Jaakkola This is my paper · ORCID
classification stat.MLcs.LGq-bio.QM
keywords discretecontinuousmodelsmultimodalco-designdatadfmsflow-based
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Combining discrete and continuous data is an important capability for generative models. We present Discrete Flow Models (DFMs), a new flow-based model of discrete data that provides the missing link in enabling flow-based generative models to be applied to multimodal continuous and discrete data problems. Our key insight is that the discrete equivalent of continuous space flow matching can be realized using Continuous Time Markov Chains. DFMs benefit from a simple derivation that includes discrete diffusion models as a specific instance while allowing improved performance over existing diffusion-based approaches. We utilize our DFMs method to build a multimodal flow-based modeling framework. We apply this capability to the task of protein co-design, wherein we learn a model for jointly generating protein structure and sequence. Our approach achieves state-of-the-art co-design performance while allowing the same multimodal model to be used for flexible generation of the sequence or structure.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 17 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Hacking Generative Perplexity: Why Unconditional Text Evaluation Needs Distributional Metrics

    cs.CL 2026-06 conditional novelty 7.0 of 10

    Zero-parameter naive samplers achieve state-of-the-art generative perplexity while producing incoherent text, proving the metric is unsound; distributional divergences like MAUVE and energy distance correctly rank the...

  2. Neuro-Symbolic ODE Discovery with Latent Grammar Flow

    cs.LG 2026-04 unverdicted novelty 7.0 of 10

    Latent Grammar Flow embeds grammar-based ODE representations into a discrete latent space with a behavioural loss and samples candidate equations via discrete flow to fit observed data.

  3. Design-CP: Context Parallelism for Design of Protein Nanoparticles

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Context-parallel inference for RFdiffusion 3 enables end-to-end all-atom design of large symmetric protein nanoparticles on multi-GPU hardware without retraining.

  4. HD-Prot: A Protein Language Model for Joint Sequence-Structure Modeling with Continuous Structure Tokens

    cs.CE 2025-12 conditional novelty 6.0 of 10

    HD-Prot shows that a protein language model can jointly generate sequences and structures using continuous structure tokens instead of quantized tokens, reaching competitive performance on four protein design tasks.

  5. FLOWR.root: A flow matching based foundation model for joint multi-purpose structure-aware 3D ligand generation and affinity prediction

    q-bio.BM 2025-10 conditional novelty 6.0 of 10

    A flow-matching model jointly generates pocket-aware 3D ligands and predicts their binding affinities, reporting state-of-the-art generation and competitive affinity accuracy with a speed advantage.

  6. Fine-Tuning Masked Diffusion for Provable Self-Correction

    cs.LG 2025-10 conditional novelty 6.0 of 10

    PRISM fine-tunes any masked diffusion model with a binary-cross-entropy loss so its new head provably estimates per-token quality p(x_i=y_i|y⊕m_i) and can remask low-quality tokens at inference.

  7. Any-Order Flexible Length Masked Diffusion

    cs.LG 2025-08 conditional novelty 6.0 of 10

    FlexMDM is a discrete diffusion model that provably supports any-order generation over variable-length sequences by learning an insertion expectation alongside the unmasking posterior, validated by length-fidelity, ma...

  8. Flexible Language Modeling in Continuous Space with Transformer-based Autoregressive Flows

    cs.LG 2025-07 conditional novelty 6.0 of 10

    TarFlowLM models language in a continuous latent space with transformer-based autoregressive normalizing flows, using mixture-CDF and Rosenblatt couplings, and reports competitive NELBO on TEXT8 and OpenWebText.

  9. Corrector Sampling in Language Models

    cs.LG 2025-06 conditional novelty 6.0 of 10

    A training and sampling method that lets autoregressive LLMs resample earlier tokens in a small window, improving reasoning and coding benchmark scores by about 10% relative after a 100B-token fine-tuning.

  10. Accelerated Sampling from Masked Diffusion Models via Entropy Bounded Unmasking

    cs.LG 2025-05 conditional novelty 6.0 of 10

    EB-Sampler dynamically unmasks multiple low-entropy tokens per function evaluation, accelerating masked diffusion model sampling by 2-3x with negligible accuracy loss.

  11. Applications of Modular Co-Design for De Novo 3D Molecule Generation

    cs.LG 2025-05 conditional novelty 6.0 of 10

    A new transformer architecture with joint continuous and discrete denoising improves 3D molecule generation and moves generated structures closer to low-energy physical minima.

  12. Flow Matching for Collaborative Filtering

    cs.IR 2025-02 conditional novelty 6.0 of 10

    FlowCF applies flow matching with a behavior-guided prior and a discrete flow framework to collaborative filtering, achieving state-of-the-art top-N recommendation accuracy with two-step inference.

  13. Provable diffusion-based posterior sampling for linear inverse problems via DDIM

    cs.LG 2026-07 reject novelty 5.0 of 10

    A SVD-based, coordinate-wise DDIM sampler is claimed to asymptotically sample from the posterior for noisy linear inverse problems, but the proof's posterior identification step does not follow from the stated updates.

  14. Energy-Based Flow Matching for Generating 3D Molecular Structure

    cs.LG 2025-08 conditional novelty 5.0 of 10

    IDFlow trains a flow matching network to refine its own predicted 3D molecular structure, improving docking and protein backbone generation over HarmonicFlow and FrameFlow baselines.

  15. MolFORM: Multi-modal Flow Matching for Structure-Based Drug Design

    cs.CE 2025-07 conditional novelty 5.0 of 10

    A flow-matching model with direct preference optimization fine-tuning generates protein-binding molecules faster than diffusion baselines, with improved docking scores on the CrossDocked2020 benchmark.

  16. TABASCO: A Fast, Simplified Model for Molecular Generation with Improved Physical Quality

    cs.LG 2025-07 conditional novelty 5.0 of 10

    TABASCO achieves 0.92 PoseBusters validity on GEOM-Drugs with a 59M-parameter non-equivariant transformer, no bond modeling, and post-hoc RDKit bond recovery, while sampling about 10x faster than SemlaFlow.

  17. AffinityFlow: Guided Flows for Antibody Affinity Maturation

    cs.LG 2025-02 reject novelty 5.0 of 10

    AffinityFlow guides AlphaFlow structure generation toward low Rosetta binding energy, then inverse-folds the structures to propose antibody mutations, and reports top scores on a computational affinity maturation benchmark.

Pith tools