DRelay: Global Draft Context for Prefix-Aware Parallel Speculative Decoding Repair

arXiv:2610.01439 · cs.AI · Submitted 2026-10-01 · Read on arXiv

cs.AI

Submitted: 2026-10-01

Updated: 2026-10-01

Code: https://github.com/tatsu-lab/stanford_alpaca

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Speculative Decoding — 1× in this paper · explained in 2 episodes

Sources

• Program Synthesis with Large Language Models
• Accelerating Large Language Model Decoding with Speculative Sampling
• Evaluating Large Language Models Trained on Code
• DSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation
• Training Verifiers to Solve Math Word Problems
• Verification-Aware Training for Speculative Decoding
• Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding
• Draft-OPD: On-Policy Distillation for Speculative Draft Models
• SpecForge: A Flexible and Efficient Open-Source Training Framework for Speculative Decoding
• Let's Verify Step by Step
• ReTrace: Rejected-Trajectory Conditioning for Speculative Decoding
• AdaFlash: Adaptive Speculative Decoding via On-Policy Distilled Diffusion Drafters
• TAPS: Target-Aware Prefix Tree Selection for Diffusion-Drafted Speculative Decoding
• Online Draft Co-Training for Speculative Decoding in Large-Scale, Long-Context RL Post-Training
• D-PACE: Dynamic Position-Aware Cross-Entropy for Parallel Speculative Drafting
• The Perfect Blend: Redefining RLHF with Mixture of Judges
• Qwen3 Technical Report
• DFlare: Scaling Up Draft Capacity for Block Diffusion Speculative Decoding
• DFlow: Enabling Verifier Information Flow in Block Diffusion Speculative Decoding

Related papers

• 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
• Agents' Last Exam
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems