Strong Drafts Need Compact Memories: Long-Context Speculative Decoding with Compressed KV Cache

August 31, 2026 ยท Grace Period ยท ๐Ÿ› EMNLP 2026 Findings

โณ Grace Period
This paper is less than 90 days old. We give authors time to release their code before passing judgment.
Authors Tong Yuan, Chengxi Liao, Zeyi Wen arXiv ID 2608.30252 Category cs.LG: Machine Learning Citations 0 Venue EMNLP 2026 Findings
Abstract
Long-context LLM applications such as document summarization and multi-turn agents require generation from prefixes spanning tens of thousands of tokens, making decoding latency a major bottleneck. Speculative decoding (SD) reduces latency without changing model outputs, but its speedup depends on both accepted draft tokens and draft-step latency: Lightweight drafts are fast but lack the capacity to capture long-range dependencies, whereas strong independent drafts recover acceptance but incur growing KV-access cost at long prefixes. We introduce memory-augmented drafting for long-context SD, equipping a strong independent draft with compressed draft-side KV memory: A lightweight adaptor constructs and incrementally updates this memory to retain distant information and exact recent context. The target verifier retains its full KV cache and applies the standard accept/reject rule, preserving SD's lossless guarantee. Experiments on Llama~3.1-8B and 70B targets at prefix lengths up to 32K show that our method reduces draft-side memory by over 70%. It achieves speedups of up to 2.08x and 3.33x , respectively, over autoregressive decoding.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

๐Ÿ“œ Similar Papers

In the same crypt โ€” Machine Learning