Latent Refinement Decoding: Enhancing Diffusion-Based Language Models by Refining Belief States
cs.CL
Submitted: 2025-10-13
Updated: 2026-11-06
Terminology
Sources
- GPT-4 Technical Report
- Program Synthesis with Large Language Models
- Accelerated Sampling from Masked Diffusion Models via Entropy Bounded Unmasking
- Evaluating Large Language Models Trained on Code
- Training Verifiers to Solve Math Word Problems
- Lipschitz Normalization for Self-Attention Layers with Application to Graph Neural Networks
- Advancing Sequential Numerical Prediction in Autoregressive Models
- Training Large Language Models to Reason in a Continuous Latent Space
- SpecFormer: Guarding Vision Transformer Robustness via Maximum Singular Value Penalization
- Accelerating Diffusion LLMs via Adaptive Parallel Decoding
- Mercury: Ultra-Fast Language Models Based on Diffusion
- Breaking AR's Sampling Bottleneck: Provable Acceleration via Diffusion Language Models
- Beyond Fixed: Training-Free Variable-Length Denoising for Diffusion Large Language Models
- Diffusion Language Models Know the Answer Before Decoding
- Promises and Pitfalls of Generative Masked Language Modeling: Theoretical Framework and Practical Guidelines
- dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Caching
- Plan for Speed: Dilated Scheduling for Masked Diffusion Language Models
- dKV-Cache: The Cache for Diffusion Language Models
- GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models
- Large Language Diffusion Models
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data