TDLM reuses past denoiser states to speed diffusion language models
On LLaDA the added modules are about 0.84% of parameters and generation is up to 7.15 times faster.
Stories per day · 30 days
3 RSS
Relateddiffusion-language-models4dream2inference2credit-assignment1decoding1gsm8k1math-reasoning1reinforcement-learning1schedule-distillation1temporal-self-conditioning1unmasking1
10 stories, newest first · one card per story
Stories tagged #llada
dSpark distills unmasking schedules to speed diffusion language models
On GSM8K the paper reports 10.12× and 5.12× speedups on LLaDA-8B-Instruct and Dream-7B-Instruct while keeping task quality.
DTCA assigns token-level credit for RL on diffusion language models
On LLaDA-8B-Instruct, authors report gains of up to 9.0 points on MATH500 and 5.3 on GSM8K.
Study isolates what improves learned unmasking in diffusion language models
An 856K-parameter policy on LLaDA-8B-Instruct leads the accuracy–NFE frontier on GSM8K and largely transfers to Dream-7B-Instruct.
Noise Out, Bias In: Targeted Bias Injection in Diffusion Language Models via Closed-Loop Activation Steering
A closed-loop steering attack injects targeted demographic bias into diffusion language models during denoising.
Representation-Space MMD for Diffusion Language Models
Representation-space MMD post-training improves diffusion language models without full sampling trajectories.
Pivot-SD: Efficient Self-Distillation for Masked Diffusion Language Models
Pivot-SD self-distills masked diffusion models by training only high-impact commitments, improving LLaDA-8B on math and code.
Unmask the State: When Does State Adaptation Matter for Masked Diffusion Language Models
Selective state adaptation captures most masked-diffusion gains, including 56.9% of LLaDA-8B oracle opportunity.
Register Tokens for Bounded-State Reasoning in Diffusion Language Models
Register tokens let diffusion language models like LLaDA and Dream carry reasoning state across cleared chunks, gaining up to 19.5 points on code.
LLaDA-UI: Bringing Block-wise Diffusion to Vision-Language GUI Agents
LLaDA-UI, a 16.7B block-wise diffusion vision-language GUI agent, outperforms Qwen2.5-VL-7B and beats Qwen3-VL-8B on four of six GUI benchmarks.