News for hackers. Stuff that matters.
Re:Δ-Mem: Efficient Online Memory for Large Language Models by jandrese on Saturday May 16, 2026 @09:48AM (#48160253)
So instead of a FIFO approach to memory management it instead continually degrades the existing data the more you put in? Details start getting lost or mangled more and more over time?
Re:Δ-Mem: Efficient Online Memory for Large Language Models by trollbridge on Saturday May 16, 2026 @12:27PM (#48161595)
That’s basically what happens.

As you hit the limits and try to compact the context, etc., things get more erratic.