Simple Diffusion Language Models Are More Effective Few-Step Generators Than Reported
cs.CL, cs.LG
Submitted: 2026-09-27
Updated: 2026-09-27
Code: https://github.com/shasanamin/more-effective-dlms
Project page: https://skylion007.github.io/OpenWebTextCorpus
Terminology
Sources
- Latent-Kernel Discrete Flow Maps for Few-Step Generation
- Enabling Approximate Joint Sampling in Diffusion LMs
- Evaluating Large Language Models Trained on Code
- Training Verifiers to Solve Math Word Problems
- Uniform Diffusion Models Revisited: Leave-One-Out Denoiser and Absorbing State Reformulation
- Error Bounds and Optimal Schedules for Masked Diffusions with Factorized Approximations
- Attention-Discounted Adaptive Sampler for Masked Diffusion Language Models
- CaRE Compute-aware Remasking Evaluation Protocol for Masked Diffusion Language Models
- The data geometry of masking diffusion: Certified-optimal schedules via unmasking growth complexity
- Qwen3 Technical Report
- Dream 7B: Diffusion Large Language Models
- Generation Order and Parallel Decoding in Masked Diffusion Models: An Information-Theoretic Perspective
- NoveltyBench: Evaluating Language Models for Humanlike Diversity
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering