Train the Sampler You Run: Few-Step Masked Diffusion Language Models via Mixture-Likelihood Fine-Tuning
cs.LG, cs.CL
Submitted: 2026-11-02
Updated: 2026-11-02
Terminology
Sources
- Simple Diffusion Language Models Are More Effective Few-Step Generators Than Reported
- Accelerated Sampling from Masked Diffusion Models via Entropy Bounded Unmasking
- Optimal Inference Schedules for Masked Diffusion Models
- Beyond Autoregression: Fast LLMs via Self-Distillation Through Time
- Learning Generation Orders for Masked Discrete Diffusion Models via Variational Inference
- Alpha Diffusion Language Models: Factorization Alone Is Not the Problem
- MDPO: Overcoming the Training-Inference Divide of Masked Diffusion Language Models
- E-MoE: Enhanced Mixture-of-Experts for Non-Factorized Diffusion Language Models
- Tensor-Train Joint Modeling for Few-Step Discrete Diffusion
- Randomization Beyond Deterministic Adaptivity in Parallel Sampling
- On Trajectory-Aware Training for Masked Diffusion Language Models
- DUEL: Exact Likelihood Reshapes Masked Diffusion Scaling Laws and Beyond
- KL Before and After Mixing: A Round-Complexity Separation in Parallel Sampling
- Few-Step Diffusion Language Models via Trajectory Self-Distillation
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis