Self-conditioned flow language models solve fixed-point iterations, enabling fixed-point flow maps that distill into FMLM* which outperforms SOTA in few-step generation on OpenWebText.
Categorical flow maps.arXiv preprint arXiv:2602.12233
14 Pith papers cite this work. Polarity classification is still indexing.
14
Pith papers citing it
citation-role summary
background 4
citation-polarity summary
years
2026 14roles
background 4polarities
background 4representative citing papers
MLFMs combine masking with continuous flows to scale flow-based language models to reasoning and instruction-following tasks on GSM8K and MT-Bench.
Continuous language diffusion works by entering high-margin decoder basins where frozen T5 embeddings recover 93-96% of native decisions and linear readouts reach 97.9% agreement, implying models should be evaluated as representation-decoder systems.
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.
A 130M-parameter continuous bitstream diffusion model with entropy-gated Langevin sampling achieves GenPPL 59.76 on LM1B and 27.06 on OWT, closing the gap to autoregressive models at matched entropy with 256 NFEs.
FMLM+ with Posterior Refinement bridges masked diffusion and flow map models to match discrete baseline quality in language generation using 32x fewer neural function evaluations via posterior scoring and refinement.
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 them below trained models.
DeCAF distills all-atom cofolding diffusion models into few-step flow maps, showing improved or matched accuracy on protein-ligand tasks with 5x fewer inference steps.
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 applies continuous-time flow matching in embedding space for language generation and reports outperforming prior discrete and continuous diffusion language models with fewer 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.
Continuous flows on token embeddings with flow-map distillation produce one-step language models whose quality exceeds recent 8-step discrete diffusion baselines on LM1B and OpenWebText.
citing papers explorer
-
Self-conditioned Flow Map Language Models via Fixed-point Flows
Self-conditioned flow language models solve fixed-point iterations, enabling fixed-point flow maps that distill into FMLM* which outperforms SOTA in few-step generation on OpenWebText.
-
Masked Language Flow Models
MLFMs combine masking with continuous flows to scale flow-based language models to reasoning and instruction-following tasks on GSM8K and MT-Bench.
-
Continuous Language Diffusion as a Decoder-Interface Problem
Continuous language diffusion works by entering high-margin decoder basins where frozen T5 embeddings recover 93-96% of native decisions and linear readouts reach 97.9% agreement, implying models should be evaluated as representation-decoder systems.
-
Drifting Objectives for Refining Discrete Diffusion Language Models
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.
-
Sampling from Flow Language Models via Marginal-Conditioned Bridges
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.
-
Towards Closing the Autoregressive Gap in Language Modeling via Entropy-Gated Continuous Bitstream Diffusion
A 130M-parameter continuous bitstream diffusion model with entropy-gated Langevin sampling achieves GenPPL 59.76 on LM1B and 27.06 on OWT, closing the gap to autoregressive models at matched entropy with 256 NFEs.
-
Posterior Refinement: Fast Language Generation via Any-Order Flow Maps
FMLM+ with Posterior Refinement bridges masked diffusion and flow map models to match discrete baseline quality in language generation using 32x fewer neural function evaluations via posterior scoring and refinement.
-
Hacking Generative Perplexity: Why Unconditional Text Evaluation Needs Distributional Metrics
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 them below trained models.
-
Few-step Cofolding with All-Atom Flow Maps
DeCAF distills all-atom cofolding diffusion models into few-step flow maps, showing improved or matched accuracy on protein-ligand tasks with 5x fewer inference steps.
-
Continuous Diffusion Scales Competitively with Discrete Diffusion for Language
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: Embedded Language Flows
ELF applies continuous-time flow matching in embedding space for language generation and reports outperforming prior discrete and continuous diffusion language models with fewer steps.
-
How to Train Your Latent Diffusion Language Model Jointly With the Latent Space
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 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.
-
Flow Map Language Models: One-step Language Modeling via Continuous Denoising
Continuous flows on token embeddings with flow-map distillation produce one-step language models whose quality exceeds recent 8-step discrete diffusion baselines on LM1B and OpenWebText.