LLaDA is a scalable diffusion-based language model that matches autoregressive LLMs like LLaMA3 8B on tasks and surpasses GPT-4o on reversal poem completion.
hub
Diffusionbert: Improving generative masked language models with diffusion models
13 Pith papers cite this work. Polarity classification is still indexing.
13
Pith papers citing it
hub tools
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
Set diffusion factorizes likelihood over arbitrary token sets and uses a set-causal diffusion architecture to support KV caching and any-order decoding, yielding improved speed-quality tradeoffs versus prior diffusion LMs.
NI Sampling accelerates discrete diffusion language models up to 14.3 times by training a neural indicator to select which tokens to sample at each step using a trajectory-preserving objective.
Early and late denoising steps in masked diffusion LMs are robust to smaller-model replacement, enabling 17% FLOPs reduction with modest generative quality loss.
Fast-dLLM adds reusable KV cache blocks and selective parallel decoding to diffusion LLMs, closing most of the speed gap with autoregressive models without retraining.
Absorbing discrete diffusion models the conditional distributions of clean data; reparameterizing yields a time-independent RADD that unifies with AO-ARMs and reaches SOTA perplexity among diffusion models on zero-shot language benchmarks.
In diffusion language models, coarse linguistic labels stabilize earlier than exact token identity, uncertainty tracks correctness, and mid-trajectory states are most sensitive to perturbations.
Efficient-DLM converts AR models to dLMs via block-wise causal attention and position-dependent masking, yielding higher accuracy and 2.7-4.5x throughput than Dream 7B and Qwen3 4B.
AsyncVLA adds asynchronous flow matching and a confidence rater to VLA models so they can generate actions on flexible schedules and selectively refine low-confidence tokens before execution.
LLaDA-V is a diffusion-based multimodal large language model that reaches competitive or state-of-the-art results on visual instruction tasks while using a non-autoregressive architecture.
GCDance is a text-and-music-conditioned diffusion framework that generates genre-consistent 3D dance sequences and reports better results than prior methods on FineDance and AIST++.
VADD augments masked diffusion models with an auxiliary recognition model and variational inference to implicitly model inter-dimensional correlations, yielding higher-quality samples than standard MDMs at low denoising step counts on toy data, images, and text.
Review of neural scaling laws and their relation to constraints and inductive biases when applying machine learning to physics problems.
citing papers explorer
-
Statistical Properties of Training & Generalization
Review of neural scaling laws and their relation to constraints and inductive biases when applying machine learning to physics problems.