paper-with-me

Papers

Cacheback: Speculative Decoding With Nothing But Cache

2025-11-15 · Zhiyao Ma, In Gim, Lin Zhong arxiv

We present Cacheback Decoding, a training-free and model-agnostic speculative decoding method that exploits the locality in language to accelerate Large Language Model (LLM) inference. Cacheback leverages only Least Recently Used (LRU) cache tables of token n-grams to generate draft sequences. Cacheback achieves state-of-the-art performance among comparable methods despite its minimalist design, and its simplicity allows easy integration into existing systems. Cacheback also shows potential for fast adaptation to new domains.

📄 PDF Abstract BibTeX arXiv:2511.21699

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache

2025-02-05 · Rishabh Tiwari, Haocheng Xi, Aditya Tomar, Coleman Hooper 외

Large Language Models (LLMs) are increasingly being deployed on edge devices for long-context settings, creating a growing need for fast and efficient long-context inference. In these scenarios, the Key-Value (KV) cache …

GPU

Hybrid Verified Decoding: Learning to Allocate Verification in Speculative Decoding

2026-05-31 · Xin Su, Dawid Majchrowski, Fangyuan Yu, Vanshil Atul Shah 외 arxiv

Large Language Model (LLM) generation remains expensive because autoregressive decoding calls the model once for each new token. Speculative decoding reduces this cost by drafting multiple tokens and verifying them with …

Vegas: Self-Speculative Decoding with Verification-Guided Sparse Attention

2026-02-06 · Yikang Yue, Yuqi Xue, Jian Huang arxiv

Long-context large language model (LLM) inference has become the norm for today's AI applications. However, it is severely bottlenecked by the increasing memory demands of its KV cache. Previous works have shown that sel…

BudgetDraft: Acceptance-Aware Multi-View Training for Sparse-KV Speculative Decoding

2026-05-29 · Liang He, Jingbo Wen, Qishi Zhan, Yixiong Chen 외 arxiv

Speculative decoding speeds up autoregressive decoding by using a drafter to propose multiple tokens that a verifier validates in parallel. In resource-constrained deployments, the drafter uses a sparse KV cache to limit…

Nightjar: Dynamic Adaptive Speculative Decoding for Large Language Models Serving

2025-12-27 · Rui Li, Zhaoning Zhang, Libo Zhang, Huaimin Wang 외 arxiv

Speculative decoding (SD) accelerates LLM inference by verifying draft tokens in parallel. However, this method presents a critical trade-off: it improves throughput in low-load, memory-bound systems but degrades perform…