# DSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation

arXiv: https://arxiv.org/abs/2607.05147 · HTML: https://mlsanity.com/paper/2607.05147/

## Key facts

- Cited by 45 papers in the map (data through 14 Oct 2026).
- Placed in Speculative decoding, one of 12 directions in Efficient LLM inference and serving.
- Builds on 60 of its 64 references that are in the map.
- First posted on arXiv 6 Jul 2026.

## Placed in

- Efficient LLM inference and serving / [Speculative decoding](https://mlsanity.com/inference/speculative-decoding/)

## Builds on (most cited first, 30 of 60)

- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [CARLA: An Open Urban Driving Simulator](https://mlsanity.com/paper/1711.03938/) (2017)
- [World Models](https://mlsanity.com/paper/1803.10122/) (2018)
- [Planning-oriented Autonomous Driving](https://mlsanity.com/paper/2212.10156/) (2022)
- [GAIA-1: A Generative World Model for Autonomous Driving](https://mlsanity.com/paper/2309.17080/) (2023)
- [VAD: Vectorized Scene Representation for Efficient Autonomous Driving](https://mlsanity.com/paper/2303.12077/) (2023)
- [TransFuser: Imitation with Transformer-Based Sensor Fusion for Autonomous Driving](https://mlsanity.com/paper/2205.15997/) (2022)
- [World Action Models are Zero-shot Policies](https://mlsanity.com/paper/2602.15922/) (2026)
- [ST-P3: End-to-end Vision-based Autonomous Driving via Spatial-Temporal Feature Learning](https://mlsanity.com/paper/2207.07601/) (2022)
- [NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking](https://mlsanity.com/paper/2406.15349/) (2024)
- [Fast-WAM: Do World Action Models Need Test-time Future Imagination?](https://mlsanity.com/paper/2603.16666/) (2026)
- [Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllability](https://mlsanity.com/paper/2405.17398/) (2024)
- [DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving](https://mlsanity.com/paper/2411.15139/) (2024)
- [Causal World Modeling for Robot Control](https://mlsanity.com/paper/2601.21998/) (2026)
- [Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving](https://mlsanity.com/paper/2311.17918/) (2023)
- [Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning](https://mlsanity.com/paper/2601.16163/) (2026)
- [MagicDrive: Street View Generation with Diverse 3D Geometry Control](https://mlsanity.com/paper/2310.02601/) (2023)
- [Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Driving](https://mlsanity.com/paper/2406.03877/) (2024)
- [Unified World Models: Coupling Video and Action Diffusion for Pretraining on Large Robotic Datasets](https://mlsanity.com/paper/2504.02792/) (2025)
- [OccWorld: Learning a 3D Occupancy World Model for Autonomous Driving](https://mlsanity.com/paper/2311.16038/) (2023)
- [GenAD: Generative End-to-End Autonomous Driving](https://mlsanity.com/paper/2402.11502/) (2024)
- [Scene as Occupancy](https://mlsanity.com/paper/2306.02851/) (2023)
- [Unified Video Action Model](https://mlsanity.com/paper/2503.00200/) (2025)
- [ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Driving](https://mlsanity.com/paper/2506.08052/) (2025)
- [DriveDreamer-2: LLM-Enhanced World Models for Diverse Driving Video Generation](https://mlsanity.com/paper/2403.06845/) (2024)
- [Deep Patch Visual Odometry](https://mlsanity.com/paper/2208.04726/) (2022)
- [GAIA-2: A Controllable Multi-View Generative World Model for Autonomous Driving](https://mlsanity.com/paper/2503.20523/) (2025)
- [DreamGen: Unlocking Generalization in Robot Learning through Video World Models](https://mlsanity.com/paper/2505.12705/) (2025)

## Built on by (most cited first, 30 of 45)

- [SEED: Self-Evolving On-Policy Distillation for Agentic Reinforcement Learning](https://mlsanity.com/paper/2607.14777/) (2026)
- [xPress: Parallel Refinement for Diffusion Drafters in Speculative Decoding](https://mlsanity.com/paper/2608.02438/) (2026)
- [DARTree: Speculative Diffusion Decoding with Autoregressive Draft Trees](https://mlsanity.com/paper/2608.13524/) (2026)
- [DeLS-Spec: Decoupled Long-Short Contexts for Parallel Speculative Drafting](https://mlsanity.com/paper/2607.07409/) (2026)
- [AdaFlash: Adaptive Speculative Decoding via On-Policy Distilled Diffusion Drafters](https://mlsanity.com/paper/2607.19223/) (2026)
- [AngelSpec: Towards Real-World High Performance Inference with Speculative Decoding](https://mlsanity.com/paper/2607.25852/) (2026)
- [LiLiCorr: Lightweight Likelihood Correlation of Parallel Drafts for Speculative Decoding](https://mlsanity.com/paper/2608.20530/) (2026)
- [A Practical Investigation of Training-free Relaxed Speculative Decoding](https://mlsanity.com/paper/2607.08690/) (2026)
- [From Chains to Trees: Parent-Conditioned Drafting for Semi-Autoregressive Speculative Decoding](https://mlsanity.com/paper/2608.02123/) (2026)
- [Predict, Don't Iterate: Efficient Adaptive-Length Infilling for Diffusion Language Models](https://mlsanity.com/paper/2609.02108/) (2026)
- [Online Draft Co-Training for Speculative Decoding in Large-Scale, Long-Context RL Post-Training](https://mlsanity.com/paper/2609.07108/) (2026)
- [DraftExpert: Expansion-Aware Self-Speculative Decoding for End-Device MoE Inference](https://mlsanity.com/paper/2607.24434/) (2026)
- [ReTrace: Rejected-Trajectory Conditioning for Speculative Decoding](https://mlsanity.com/paper/2608.29748/) (2026)
- [DFlow: Enabling Verifier Information Flow in Block Diffusion Speculative Decoding](https://mlsanity.com/paper/2609.06498/) (2026)
- [LLaDA MoE v2: Scaling Mixture-of-Experts Diffusion Language Models](https://mlsanity.com/paper/2608.03457/) (2026)
- [DBLAST: Dependent Block Drafting for Stochastic Speculative Decoding](https://mlsanity.com/paper/2608.05448/) (2026)
- [Training Agents to Evolve with Their Harness: TaoLive Digital Avatar Agent Technical Report](https://mlsanity.com/paper/2608.15763/) (2026)
- [Carryover Drafting: Recycling Rejected States for Speculative Decoding](https://mlsanity.com/paper/2609.14717/) (2026)
- [TreeSpark: Calibrated, Load-Adaptive Draft Trees for Semi-Autoregressive Speculative Decoding](https://mlsanity.com/paper/2609.22098/) (2026)
- [Draft in Parallel, Condition Through Depth: Adjacent Causal Injection for Speculative Decoding](https://mlsanity.com/paper/2609.36173/) (2026)
- [From Expert Reduction to Behavioral Divergence: Tracing Numerical State through Sparse MoE Inference](https://mlsanity.com/paper/2607.28097/) (2026)
- [K-EXAONE 2.0 Technical Report](https://mlsanity.com/paper/2608.04505/) (2026)
- [From Positionwise Confidence to Prefix Scheduling: Verifier Skipping in Speculative Decoding](https://mlsanity.com/paper/2608.14787/) (2026)
- [Is Multimodal Speculative Decoding Ready for Diffusion-Based Parallel Drafting? A Survey and Empirical Diagnosis](https://mlsanity.com/paper/2608.20743/) (2026)
- [NCP-ArchPreview Technical Report: Moving towards Latent Space Language Models through Next Concept Prediction](https://mlsanity.com/paper/2609.10715/) (2026)
- [Reason What Matters: Retrieval-Grounded Reasoning for Universal Multimodal Embeddings](https://mlsanity.com/paper/2609.15296/) (2026)
- [Adapting Tree-Structured Speculative Decoding to DeepSeek-V4 for Efficient Inference](https://mlsanity.com/paper/2609.24698/) (2026)
- [From Position Risks to Block Survival: Faster Generation for Diffusion Language Models](https://mlsanity.com/paper/2609.33390/) (2026)
- [Tsubame: Tree Replay for Diffusion-Based Speculative Decoding](https://mlsanity.com/paper/2609.33652/) (2026)
- [Reciprocal Guidance: Orchestrating Draft and Verify Budgets for Advancing the Diffusion-AR Self-Speculation Frontier](https://mlsanity.com/paper/2609.34388/) (2026)

## Abstract (arXiv)

Speculative decoding accelerates Large Language Model (LLM) inference by decoupling draft generation from target verification. While recent parallel drafters efficiently propose long token sequences in a single forward pass, they suffer from rapid acceptance decay due to a lack of inter-token dependencies. Furthermore, indiscriminately verifying these extended blocks wastes critical batch capacity on tokens with high rejection risks, severely degrading throughput in high-concurrency serving systems. We introduce DSpark, a speculative decoding framework that unifies high-throughput parallel generation with adaptive, load-aware verification. To maintain draft quality, DSpark utilizes a semi-autoregressive architecture, coupling a parallel backbone with a lightweight sequential module, to introduce intra-block dependency modeling and mitigate suffix decay. To optimize system efficiency, DSpark employs confidence-scheduled verification, dynamically tailoring the verification length for each request based on estimated prefix survival probabilities and engine-specific throughput profiles. On offline benchmarks across diverse domains, DSpark substantially improves the accepted length over state-of-the-art autoregressive and parallel drafters. When deployed within the DeepSeek-V4 serving system under live user traffic, DSpark successfully mitigates verification waste. Compared to the established production baseline (MTP-1), DSpark accelerates per-user generation speeds by 60 to 85 percent at matched throughput levels. More importantly, by preventing severe throughput degradation under strict interactivity constraints, it enables performance tiers that were previously unattainable, shifting the Pareto frontier of our serving system.