🧠 Memory Management - hello · Scour

A new era for memory-management maintainership 🧱Slab Allocation

·5d

LKV: End-to-End Learning of Head-wise Budgets and Token Selection for LLM KV Cache Eviction ⏪Deoptimization

When Does Value-Aware KV Eviction Help? A Fixed-Contract Diagnostic for Non-Monotone Cache Compression ⏪Deoptimization

Not All Thoughts Need HBM: Semantics-Aware Memory Hierarchy for LLM Reasoning 📏Linear Types

ClickHouse/silk: Fast stackful fibers with a NUMA-aware work-stealing scheduler 📊Profiling Tools

github.com·6d·Hacker News

Shepherd: A Runtime Substrate Empowering Meta-Agents with a Formalized Execution Trace 🕵️Jaeger

RateQuant: Optimal Mixed-Precision KV Cache Quantization via Rate-Distortion Theory 🚀Superoptimization

I built a baremetal RustOS (O(1)allocator NVMeDMA ZeroTrust sandbox for AIAgents 🐧Rust for Linux

github.com·3d·Hacker News

Make Each Token Count: Towards Improving Long-Context Performance with KV Cache Eviction 🧱Slab Allocation

MemPrivacy: Privacy-Preserving Personalized Memory Management for Edge-Cloud Agents 🔗Intrusive Containers

ipc_shared_ptr: A Publish/Subscribe-Aware Smart Pointer for Cross-Process Object Lifetime Management 🗺️Memory-Mapped Queues

Efficient Serving for Dynamic Agent Workflows with Prediction-based KV-Cache Management ⏰Timely Dataflow

When Quantization Is Free: An int4 KV Cache That Outruns fp16 on Apple Silicon 🏗Computer Architecture

Tutti: Making SSD-Backed KV Cache Practical for Long-Context LLM Serving 🎴TAO

eOptShrinkQ: Near-Lossless KV Cache Compression Through Optimal Spectral Denoising and Quantization ⏲️TimescaleDB Compression

HeadQ: Model-Visible Distortion and Score-Space Correction for KV-Cache Quantization 🌊Memory Bandwidth

Log in to enable infinite scrolling