12 papers
ImpactHO: Importance-Aware KV Cache Transfer for Multi-User Edge LLM Handover
Minwoo Kim, Soochang Song, Namyoon Lee +2
Edge LLMs must preserve inference continuity when a user hands over between edge nodes, requiring key-value (KV) cache transfer to the target node. However, simultaneous handovers…
Collaborative Edge-to-Server Inference for Vision-Language Models
Soochang Song, Yongjune Kim
We propose a collaborative edge-to-server inference framework for vision-language models (VLMs) that reduces communication cost while maintaining inference accuracy. In typical dep…
Mutual Information Minimization for Side-Channel Attack Resistance via Optimal Noise Injection
Jiheon Woo, Donggyun Ryu, Daewon Seo +4
Side-channel attacks (SCAs) pose a serious threat to system security by extracting secret keys through physical leakages such as power consumption, timing variations, and electroma…
FibQuant: Universal Vector Quantization for Random-Access KV-Cache Compression
Namyoon Lee, Yongjune Kim
Long-context inference is increasingly a memory-traffic problem. The culprit is the key--value (KV) cache: it grows with context length, batch size, layers, and heads, and it is re…
CGF-Softmax: A Cumulant-Based Softmax Reformulation for Efficient Inference under Homomorphic Encryption
Hanjun Park, Byeongseo Min, Jiheon Woo +5
Homomorphic encryption (HE) is a prominent framework for privacy-preserving machine learning, enabling inference directly on encrypted data. However, evaluating softmax, a core com…
A Survey on Robust Deep Joint Source-Channel Coding for Semantic Communications
Eunhye Hong, Taewoo Park, Yongjune Kim
Semantic communications (SCs) aim to transmit only the essential information required to perform given tasks, thereby improving communication efficiency. Deep learning-based joint…