2026
Dustin: Draft-Augmented Sparse Verification for Efficient Long-Context Generation with Speculative Decoding
ICML 2026poster
While speculative decoding improves inference throughput for multi-batch long-context Large Language Models (LLMs), its efficiency is often limited by a verification bottleneck where Key-Value (KV) cache loading dominates latency. Existing compression methods fail in this regime: static eviction inc…