Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Protection Is (Nearly) All You Need: Structural Protection Dominates Scoring in Globally Capped KV Eviction
Gabriel Garcia
We study KV cache eviction under a shared globally capped decode-time harness. Seven policies (LRU, H2O, SnapKV, StreamingLLM, Ada-KV, QUEST, Random) share a prompt-boundary vulner…
cs.LG2026
No Free Swap: Protocol-Dependent Layer Redundancy in Transformers
Gabriel Garcia
When researchers ask whether two transformer layers are "equivalent" for compression, they often conflate distinct tests. Replacement asks whether one layer's map can substitute fo…