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.
International Conference on Learning Representations , year=
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5verdicts
UNVERDICTED 5representative citing papers
RDM trains one-step generators via MMD on large batches and multi-encoder representations, achieving SOTA SW_r14 of 1.30 on ImageNet and distilling FLUX.2 to one-step with gains on GenEval and PickScore.
Pepti-drift performs a single antigen-conditioned drift in peptide latent space to produce valid, diverse peptides with reduced toxicity and high efficiency compared to prior methods.
LineageFlow generates family-aware protein sequences via flow matching from ancestral sequence reconstruction priors, achieving near-natural family validity and improved structural confidence with a rerouting technique for guided sampling.
Hallucinations in diffusion models are driven by local intrinsic dimension instabilities on the manifold, which Intrinsic Quenching corrects by deflating it.
citing papers explorer
-
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.
-
Representation Distribution Matching for One-Step Visual Generation
RDM trains one-step generators via MMD on large batches and multi-encoder representations, achieving SOTA SW_r14 of 1.30 on ImageNet and distilling FLUX.2 to one-step with gains on GenEval and PickScore.
-
Pepti-drift: Toxicity-Repulsive Drifting for Antigen-Conditioned Discrete Peptide Generation
Pepti-drift performs a single antigen-conditioned drift in peptide latent space to produce valid, diverse peptides with reduced toxicity and high efficiency compared to prior methods.
-
LineageFlow: Flow Matching for High-Fidelity Family-Aware Protein Sequence Generation
LineageFlow generates family-aware protein sequences via flow matching from ancestral sequence reconstruction priors, achieving near-natural family validity and improved structural confidence with a rerouting technique for guided sampling.
-
Local Intrinsic Dimension Unveils Hallucinations in Diffusion Models
Hallucinations in diffusion models are driven by local intrinsic dimension instabilities on the manifold, which Intrinsic Quenching corrects by deflating it.