D3D: Decoupled Diffusion Drafting for Asynchronous Speculative Decoding
cs.AI, cs.LG
Submitted: 2026-11-04
Updated: 2026-11-04
Code: https://github.com/fudan-generative-vision/D3D
Terminology
Sources
- PARD-2: Target-Aligned Parallel Draft Model for Dual-Mode Speculative Decoding
- Distilling the Knowledge in a Neural Network
- JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Parallel Tree Drafting
- Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding
- SD$^2$: Self-Distilled Sparse Drafters
- TreeFlash: Parallel AR-Approximation for Faster Speculative Decoding
- SDFP: Speculative Decoding with FIT-Pruned Models for Training-Free and Plug-and-Play LLM Acceleration
- D-PACE: Dynamic Position-Aware Cross-Entropy for Parallel Speculative Drafting
- D^2SD: Accelerating Speculative Decoding with Dual Diffusion Draft Models
- SwiftSpec: Ultra-Low Latency LLM Decoding by Scaling Asynchronous Speculative Decoding
- DeLS-Spec: Decoupled Long-Short Contexts for Parallel Speculative Drafting
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents