Image de couverture de livre

When Drafts Evolve: Speculative Decoding Meets Online Learning

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Qian, Yu-Yang, Wu, Hao-Cong, Fu, Yichao, Zhang, Hao, Zhao, Peng
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Draft Less, Retrieve More: Hybrid Tree Construction for Speculative Decoding
par: Shen, Yuhao, et autres
Publié: (2026)

Online Speculative Decoding
par: Liu, Xiaoxuan, et autres
Publié: (2023)

Speculative Decoding with CTC-based Draft Model for LLM Inference Acceleration
par: Wen, Zhuofan, et autres
Publié: (2024)

LongSpec: Long-Context Lossless Speculative Decoding with Efficient Drafting and Verification
par: Yang, Penghui, et autres
Publié: (2025)

POSS: Position Specialist Generates Better Draft for Speculative Decoding
par: Huang, Langlin, et autres
Publié: (2025)

BubbleSpec: Turning Long-Tail Bubbles into Speculative Rollout Drafts for Synchronous Reinforcement Learning
par: Xu, Yuhang, et autres
Publié: (2026)

BudgetDraft: Acceptance-Aware Multi-View Training for Sparse-KV Speculative Decoding
par: He, Liang, et autres
Publié: (2026)

FastGRPO: Accelerating Policy Optimization via Concurrency-aware Speculative Decoding and Online Draft Learning
par: Zhang, Yizhou, et autres
Publié: (2025)

Balancing Coverage and Draft Latency in Vocabulary Trimming for Faster Speculative Decoding
par: Shoham, Ofir Ben
Publié: (2026)

ML-SpecQD: Multi-Level Speculative Decoding with Quantized Drafts
par: Georganas, Evangelos, et autres
Publié: (2025)

Training Domain Draft Models for Speculative Decoding: Best Practices and Insights
par: Hong, Fenglu, et autres
Publié: (2025)

When Linear Attention Meets Autoregressive Decoding: Towards More Effective and Efficient Linearized Large Language Models
par: You, Haoran, et autres
Publié: (2024)

Speculative Decoding Meets Quantization: Compatibility Evaluation and Hierarchical Framework Design
par: Zhang, Yudi, et autres
Publié: (2025)

Direct Alignment of Draft Model for Speculative Decoding with Chat-Fine-Tuned LLMs
par: Goel, Raghavv, et autres
Publié: (2024)

MineDraft: A Framework for Batch Parallel Speculative Decoding
par: Tang, Zhenwei, et autres
Publié: (2026)

Make Every Draft Count: Hidden State based Speculative Decoding
par: Chen, Yuetao, et autres
Publié: (2026)

D-PACE: Dynamic Position-Aware Cross-Entropy for Parallel Speculative Drafting
par: Wu, Tianyu, et autres
Publié: (2026)

When Continue Learning Meets Multimodal Large Language Model: A Survey
par: Huo, Yukang, et autres
Publié: (2025)

QSpec: Speculative Decoding with Complementary Quantization Schemes
par: Zhao, Juntao, et autres
Publié: (2024)

Cactus: Accelerating Auto-Regressive Decoding with Constrained Acceptance Speculative Sampling
par: Hao, Yongchang, et autres
Publié: (2026)

Free Draft-and-Verification: Toward Lossless Parallel Decoding for Diffusion Large Language Models
par: Wu, Shutong, et autres
Publié: (2025)

Agentic Unlearning: When LLM Agent Meets Machine Unlearning
par: Wang, Bin, et autres
Publié: (2026)

Speculative Safety-Aware Decoding
par: Wang, Xuekang, et autres
Publié: (2025)

Fail Fast, Win Big: Rethinking the Drafting Strategy in Speculative Decoding via Diffusion LLMs
par: Pan, Rui, et autres
Publié: (2025)

PRISM: Parametrically Refactoring Inference for Speculative Sampling Draft Models
par: Wang, Xuliang, et autres
Publié: (2026)

BanditSpec: Adaptive Speculative Decoding via Bandit Algorithms
par: Hou, Yunlong, et autres
Publié: (2025)

STree: Speculative Tree Decoding for Hybrid State-Space Models
par: Wu, Yangchao, et autres
Publié: (2025)

Fast Large Language Model Collaborative Decoding via Speculation
par: Fu, Jiale, et autres
Publié: (2025)

DEER: Draft with Diffusion, Verify with Autoregressive Models
par: Cheng, Zicong, et autres
Publié: (2025)

When Graph Neural Network Meets Causality: Opportunities, Methodologies and An Outlook
par: Jiang, Wenzhao, et autres
Publié: (2023)

$\texttt{SPECS}$: Faster Test-Time Scaling through Speculative Drafts
par: Cemri, Mert, et autres
Publié: (2025)

Scaling Speculative Decoding with Lookahead Reasoning
par: Fu, Yichao, et autres
Publié: (2025)

Spec-VLA: Speculative Decoding for Vision-Language-Action Models with Relaxed Acceptance
par: Wang, Songsheng, et autres
Publié: (2025)

When Correct Demonstrations Hurt: Rethinking the Role of Exemplars in In-Context Learning
par: Qiu, Chenghao, et autres
Publié: (2026)

MoESD: Unveil Speculative Decoding's Potential for Accelerating Sparse MoE
par: Huang, Zongle, et autres
Publié: (2025)

When Speculation Spills Secrets: Side Channels via Speculative Decoding In LLMs
par: Wei, Jiankun, et autres
Publié: (2024)

Exploring and Improving Drafts in Blockwise Parallel Decoding
par: Kim, Taehyeon, et autres
Publié: (2024)

Mixture of Attentions For Speculative Decoding
par: Zimmer, Matthieu, et autres
Publié: (2024)

Draft-Conditioned Constrained Decoding for Structured Generation in LLMs
par: Reddy, Avinash, et autres
Publié: (2026)

Attention Drift: What Autoregressive Speculative Decoding Models Learn
par: Eldenk, Doğaç, et autres
Publié: (2026)