With the widespread deployment of long-context large language models (LLMs), there has been a growing demand for efficient support of high-throughput inference. However, as the key-value (KV) cache expands with the sequence length, the increasing memory footprint and the need to access it for each token generation both result in low throughput when serving long-context LLMs. While various dynamic sparse attention methods have been proposed to speed up inference while maintaining generation quality, they either fail to sufficiently reduce GPU memory consumption or introduce significant decodingResearch goal: Does the ShadowKV method preserve multimodal retrieval accuracy on long-document benchmarks when the KV cache size is restricted to less than 20% of the full sequence length?Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 8.1/10.