CDM amortizes SMC inference for reward-tilted discrete diffusion by training a parameterized twist function on contrastive samples with closed-form kernels.
Mixed citations
Discrete Flow Maps
Mixed citation behavior. Most common role is background (60%).
8
Pith papers citing it
Background
60% of classified citations
abstract
The sequential nature of autoregressive next-token prediction imposes a fundamental speed limit on large language models. While continuous flow models offer a path to parallel generation, they traditionally demand expensive iterative integration. Flow Maps bypass this bottleneck by compressing generative trajectories into single-step mappings, theoretically enabling the generation of full text sequences from noise in a single forward pass. However, standard formulations rely on Euclidean regression losses that are geometrically ill-suited for discrete data. In this work, we resolve this conflict with Discrete Flow Maps, a framework that reconciles trajectory compression with the geometry of the probability simplex. We recast standard flow map training for the discrete domain, aligning the training dynamics with the discrete nature of language. Empirically, this strict geometric alignment allows our method to surpass previous state-of-the-art results in discrete flow modeling.
citation-role summary
background 4
method 1
citation-polarity summary
years
2026 8representative citing papers
TokenDrift refines discrete diffusion language models by applying anti-symmetric drifting to soft-token features during training, yielding large reductions in generation perplexity at low NFEs.
Marginal-conditioned bridges enable training-free sampling from Flow Language Models by drawing clean one-hot endpoints from factorized posteriors and using Ornstein-Uhlenbeck bridges, preserving token marginals and reducing denoising error versus conditional-mean bridges.
Count-FM is a new flow-matching method for count data based on birth-death processes that achieves better sample quality with fewer parameters than baselines on simulations and real scRNA-seq and spike-train data.
RePlaid achieves a 20x compute gap to autoregressive models, new SOTA PPL of 22.1 among continuous DLMs on OpenWebText, and competitive scaling laws by aligning architecture with modern discrete DLMs.
ELF is a continuous embedding-space flow matching model for language that stays continuous until the last step and outperforms prior discrete and continuous diffusion language models with fewer sampling steps.
Joint training of the latent space with the diffusion process produces a competitive latent diffusion language model that is faster than existing discrete and continuous diffusion baselines.
Coupling Models enable single-step discrete sequence generation via learned couplings to Gaussian latents and outperform prior one-step baselines on text perplexity, biological FBD, and image FID metrics.
citing papers explorer
-
Coupling Models for One-Step Discrete Generation
Coupling Models enable single-step discrete sequence generation via learned couplings to Gaussian latents and outperform prior one-step baselines on text perplexity, biological FBD, and image FID metrics.