PACT: Certifying What Is Written in Parallel Decoding of Diffusion Language Models
cs.CL
Submitted: 2026-11-03
Updated: 2026-11-03
Comments: 17 pages, 10 figures. Code: https://github.com/LiZaiyuan0619/PACT
Code: https://github.com/LiZaiyuan0619/PACT
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Program Synthesis with Large Language Models
- Diffusion Language Models: An Experimental Analysis
- LLaDA2.1: Speeding Up Text Diffusion via Token Editing
- Evaluating Large Language Models Trained on Code
- Optimal Inference Schedules for Masked Diffusion Models
- Training Verifiers to Solve Math Word Problems
- Locally Confident, Globally Stuck: The Quality-Exploration Dilemma in Diffusion Language Models
- Reliable Parallel Decoding in Masked Diffusion Language Models
- Fast-dLLM++: Fr'echet Profile Decoding for Faster Diffusion LLM Inference
- Adaptive Test-Time Compute Allocation for Block Diffusion Language Models in Complex Reasoning
- Dependency-Guided Parallel Decoding in Discrete Diffusion Language Models
- TACG: Trajectory-Aware Commit Gating for Diffusion Language Model Decoding
- BlockBatch: Multi-Scale Consensus Decoding for Efficient Diffusion Language Model Inference
- Qwen3 Technical Report
- Dream 7B: Diffusion Large Language Models
- Dimple: Discrete Diffusion Multimodal Large Language Model with Parallel Decoding
- Generation Order and Parallel Decoding in Masked Diffusion Models: An Information-Theoretic Perspective
- Instruction-Following Evaluation for Large Language Models
- LLaDA-MoE: A Sparse MoE Diffusion Language Model
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