OPTD: On-Policy Transition Distillation with Consistency-Guided Adaptive Compression for Few-Step Diffusion Language Models
Xiaocheng Lu, Hualei Zhang, Shuhan Guo, Jie Zhang, Xiaoyi Pang, Jian Liu, Haoxi Li, Bohai Gu, Haoxuan Che, Jingcai Guo, Song Guo
cs.CL
Submitted: 2026-08-03
Comments: 9 pages, 4 figures, 5 tables
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- On-Policy Distillation of Language Models: Learning from Self-Generated Mistakes
- Program Synthesis with Large Language Models
- Evaluating Large Language Models Trained on Code
- dParallel: Learnable Parallel Decoding for dLLMs
- SDAR: A Synergistic Diffusion-AutoRegression Paradigm for Scalable Sequence Generation
- Training Verifiers to Solve Math Word Problems
- Measuring Mathematical Problem Solving With the MATH Dataset
- CD4LM: Consistency Distillation and aDaptive Decoding for Diffusion Language Models
- Learning from the Self-future: On-policy Self-distillation for dLLMs
- Large Language Diffusion Models
- d3LLM: Ultra-Fast Diffusion LLM using Pseudo-Trajectory Distillation
- Trace-Based On-Policy Distillation for Masked Diffusion Language Models
- Data-Efficient Autoregressive-to-Diffusion Language Models via On-Policy Distillation
- Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding
- Dream 7B: Diffusion Large Language Models
- Few-Step Diffusion Language Models via Trajectory Self-Distillation
- TAD: Temporal-Aware Trajectory Self-Distillation for Fast and Accurate Diffusion LLM
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