The paper proposes GroundedCache, an evidence-validated cache router that significantly improves the safety of reusing cached semantic answers in RAG systems by requiring multiple gates to validate th…

CachePrune introduces a privacy-aware, fine-grained KV cache sharing mechanism that allows LLM inference systems to safely reuse cache entries across users' requests, significantly improving efficienc…

VisMemo is a serving system that enables approximate KV cache reuse for modern vision-language models to reduce prefill latency and memory overhead.

This paper introduces HIJACKKV, the first attack framework to exploit the new threat of KV Cache Hijacking in large language models, achieving an average success rate of 94%.

This paper presents HiState, a prefix cache management system for hybrid models serving agent workloads to reduce inference costs.

Leyline introduces a novel serving-side primitive that allows agentic LLMs to perform targeted, efficient edits to the KV cache, avoiding costly full re-prefilling after content modification.

This paper proposes a method to improve multimodal agent performance by reusing cached chunks and restoring cross-chunk binding.

CXLResume reduces checkpoint and restore time by retaining shared application state in memory.

This paper introduces CW-Ghost, a method for estimating cache line fill volume and determining helper-thread prefetching granularity based on cache capacity constraints.

This paper presents a new KVCache management layer called CacheWise that improves performance for coding agent workloads.

Improves shared-library debloating by reconstructing libraries from file ranges

gKSM is a transparent GPU memory deduplication framework for serverless LLM inference that reduces GPU memory usage and improves throughput.

This paper studies the performance gap between C and Rust code generated by a transpiler and proposes an LLM-based approach to recover the performance of translated Rust.

This paper proposes C^2, a set of techniques to improve the spatial locality of succinct tries, resulting in faster query performance and reduced memory footprint.

The paper characterizes the memory-rate trade-off for demand-private coded caching with heterogeneous file sizes and cache capacities.

The paper proposes adaptive caching for key-value (KV) caching in large language models, improving cache hit rate and reducing time to first token.