Menu

#Memory

851 posts

Feed·
20 of 851 posts
I accidentally turned LLM memory into program analysis :: pwning.systems
📰
279

I accidentally turned LLM memory into program analysis :: pwning.systems

Hacker News·20 days ago
#hi5Df3DS

Why I stopped trying to give LLM agents a better memory and instead built Lemmalog, a Datalog engine that maintains an agent's knowledge as analysis state, with provenance, retractions and incremental evaluation, plus what happened when I benchmarked it…

15s
Read More
Micron Unveils Micron Research Labs, a U.S.-Based Long-Horizon Innovation Hub to Shape the Future of Memory and AI
🖼️
0

Micron Unveils Micron Research Labs, a U.S.-Based Long-Horizon Innovation Hub to Shape the Future of Memory and AI

Hacker News·28 days ago
#fqqSLqBr

Backed by a $10 billion planned investment over the next decade, Micron Research Labs will unite academia, government, startups and industry to advance the memory and compute breakthroughs that will define the AI era BOISE, Idaho, Aug.…

15s
Read More
GitHub - fellowgeek/mcp-memory: An OKF-backed Model Context Protocol (MCP) server delivering persistent long-term memory and SQLite FTS5 search for AI agents.
📰
62

GitHub - fellowgeek/mcp-memory: An OKF-backed Model Context Protocol (MCP) server delivering persistent long-term memory and SQLite FTS5 search for AI agents.

Hacker News·about 1 month ago
#PBfgLdzd

An OKF-backed Model Context Protocol (MCP) server delivering persistent long-term memory and SQLite FTS5 search for AI agents. - fellowgeek/mcp-memory

15s
Read More
GitHub - xoreaxeaxeax/skitter-creek-bath-salts: Unlocking _everything_ on the CPU with DRAM scrambling
🖼️
687

GitHub - xoreaxeaxeax/skitter-creek-bath-salts: Unlocking _everything_ on the CPU with DRAM scrambling

Hacker News·about 1 month ago
#FvNIK1wS
#github#dram#memory#core#alias#address

Unlocking _everything_ on the CPU with DRAM scrambling - xoreaxeaxeax/skitter-creek-bath-salts

15s
Read More
Now That 2027 RAM Manufacturing Capacity Has Reportedly Been Sold Through, It's Hard To Imagine the RAMageddon Ending Any Time Soon
🖼️
516

Now That 2027 RAM Manufacturing Capacity Has Reportedly Been Sold Through, It's Hard To Imagine the RAMageddon Ending Any Time Soon

Hacker News·about 1 month ago
#20L3SybZ
#ign#capacity#memory#companies#sold#article

A new report suggests that all three memory manufacturers — Samsung, SK Hynix, and Micron — have collectively sold through their 2027 memory manufacturing capacity to AI companies.

15s
Read More
GitHub - Wyzer-Lang/wyzer: The Wyzer Programming Language
🖼️
226

GitHub - Wyzer-Lang/wyzer: The Wyzer Programming Language

Hacker News·about 1 month ago
#Y6fee50f
#github#wyzer#memory#network#programming#rust

The Wyzer Programming Language. Contribute to Wyzer-Lang/wyzer development by creating an account on GitHub.

15s
Read More
Zero-Mem: Zero-Token Memory Operations for LLM Agents
🖼️
101

Zero-Mem: Zero-Token Memory Operations for LLM Agents

Hacker News·about 1 month ago
#PTeREf0h
#arxiv#memory#zero#context#xiao#long

LLM agents need memory to act consistently over long interactions, yet many systems use additional LLM calls to operate that memory. Generating intermediate records and mediating their retrieval adds recurring token and time costs, while omitted or merged…

15s
Read More
Attention Decode on AMD MI450 GPUs: A Gluon Kernel Optimization Guide
🖼️
0

Attention Decode on AMD MI450 GPUs: A Gluon Kernel Optimization Guide

Hacker News·about 2 months ago
#pjwLwqg3
#rocm#kernel#memory#attention#mi450#wmma

Learn how to design a high-performance attention decode kernel on AMD MI450 GPUs using Gluon.

15s
Read More
A very concrete explanation of how a cache works
🖼️
0

A very concrete explanation of how a cache works

Hacker News·about 2 months ago
#yJRkK32e
#parksb#cache#memory#address#index#photo

KO | EN As technology has advanced, processor speed has risen quickly, but memory speed hasn’t kept up. No matter how fast a processor is, if memory responds slowly the whole system ends up slow. The device that addresses this is the cache.…

15s
Read More
 Memory chip boss admits RAM prices are 'abnormally high' — SK Group chairman considering building a semiconductor plant in the US to expand supply, calm ‘chipflation’
📰
0

Memory chip boss admits RAM prices are 'abnormally high' — SK Group chairman considering building a semiconductor plant in the US to expand supply, calm ‘chipflation’

SK Group Chairman Chey Tae-won said that prices for memory chips are "abnormally high" and that the industry must take steps to increase production and reduce prices.…

15s
Read More