New KV cache compaction technique cuts LLM memory 50x without accuracy loss

MIT researchers developed Attention Matching, a KV cache compaction technique that compresses LLM memory by 50x in seconds — ...
InfoQ

Busting AI Myths and Embracing Realities in Privacy & Security

Katharine Jarmul keynotes on common myths around privacy and security in AI and explores what the realities are, covering design patterns that help build more secure, more private AI systems.
IEEE

Evaluating Memory Management and Garbage Collection Algorithms with Virtual Threads in High-Concurrency Java Applications

Abstract: Garbage collection (GC) is a critical memory management mechanism within the Java Virtual Machine (JVM) responsible for automating memory allocation and reclamation. Its performance affects ...