speculative decoding

CopySpec: Accelerating LLMs with Speculative Copy-and-Paste EMNLP 2025 CORAL: Learning Consistent Representations across Multi-step Training with Lighter Speculative Drafter ACL 2025 Decoding Speculative Decoding NAACL 2025 Dovetail: A CPU/GPU Heterogeneous Speculative Decoding for LLM inference EMNLP 2025 Draft Model Knows When to Stop: Self-Verification Speculative Decoding for Long-Form Generation EMNLP 2025 DReSD: Dense Retrieval for Speculative Decoding ACL 2025 Dynamic-Width Speculative Beam Decoding for LLM Inference AAAI 2025 Efficient Inference for Large Language Model-based Generative Recommendation ICLR 2025 EMS-SD: Efficient Multi-sample Speculative Decoding for Accelerating Large Language Models NAACL 2025 Falcon: Faster and Parallel Inference of Large Language Models Through Enhanced Semi-Autoregressive Drafting and Custom-Designed Decoding Tree AAAI 2025 FastDraft: How to Train Your Draft ACL 2025 Faster and Better LLMs via Latency-Aware Test-Time Scaling EMNLP 2025 Faster Cascades via Speculative Decoding ICLR 2025 Faster In-Context Learning for LLMs via N-Gram Trie Speculative Decoding EMNLP 2025 Faster Speculative Decoding via Effective Draft Decoder with Pruned Candidate Tree ACL 2025 FastVLM: Self-Speculative Decoding for Fast Vision-Language Model Inference IJCNLP 2025 Fuzzy Speculative Decoding for a Tunable Accuracy-Runtime Tradeoff ACL 2025 Grouped Speculative Decoding for Autoregressive Image Generation ICCV 2025 Growing a Twig to Accelerate Large Vision-Language Models ICCV 2025 Gumiho: A Hybrid Architecture to Prioritize Early Tokens in Speculative Decoding ICML 2025 Hardware-Aware Parallel Prompt Decoding for Memory-Efficient Acceleration of LLM Inference EMNLP 2025 Hierarchical Speculative Decoding with Dynamic Window NAACL 2025 Judge Decoding: Faster Speculative Sampling Requires Going Beyond Model Alignment ICLR 2025 LANTERN: Accelerating Visual Autoregressive Models with Relaxed Speculative Decoding ICLR 2025 LLMs on a Budget? Say HOLA EMNLP 2025