AnchorKV: Safety-Aware KV Cache Compression via Soft Penalty with a Refusal Anchor

arXiv:2606.17872v1 Announce Type: cross Abstract: Large language models (LLMs) outperform earlier architectures on generative inference and long-context tasks, but their large size introduces significant challenges in memory usage, energy cost, and on-device deployment. Since scaling pre-trained language models improves downstream capability [zhao2023survey], the key-value (KV) cache becomes a dominant inference bottleneck. Recent KV cache compression methods [jo2025fastkv,li2024snapkv,zhou2024dynamickv] reduce this cost by retaining only a subset of attention-relevant tokens. However, while these approaches preserve accuracy on benign workloads, their compression policies either fail to defend against jailbreak attacks [jiang2024robustkv] or degrade safety alignment under aggressive eviction. We propose AnchorKV, a drop-in modification to KV cache compression that biases token retention scores away from directions in key space associated with harmful prompts. AnchorKV constructs an offline safety anchor by adapting a difference-of-means representation engineering approach [arditi2024refusal,zou2023representation] to the layer-specific key projection space used in KV caching. Based on this anchor, a soft penalty token selection rule trades a small amount of utility for substantially improved safety alignment, while reducing to the original compressor when the penalty is zero.