Forked memory pages: keeping worlds separate so old knowledge never gets overwritten
RexDouglass · x · 2026-10-01
iarthsingh shared a memory mechanism for continual learning across changing "worlds," built on forking:
- Each world keeps its own page of memory;
- When several consecutive batches look unfamiliar, a new page is opened;
- Only the current page learns, so old pages can't be overwritten — preventing cross-environment interference.
The author shows the 20 pages the system discovered, unpermuted, revealing how the model segments its environments.
More from Research
- AI Overviews cites 38% fewer self-promotional listicles, drops brands 83% of the time: data — lilyraynyc · 2026-10-01
- 98,877 historical newspaper pages (1700s-1940s) with OCR text and word boxes hit Hugging Face — vanstriendaniel · 2026-10-01
- Microsoft's new ML system forecasts power-grid damage from space weather 30-60 min ahead — LesterMackey · 2026-10-01
- Noam Brown on Dwarkesh Podcast: Agent Swarms, Alignment, and Recursive Self-Improvement — GregKamradt · 2026-10-01
- Keras 3's quiet ZeroModels project packs 100+ model families across JAX, PyTorch, TensorFlow — lmoroney · 2026-10-01
- Initial probing of Opus suggests its time estimates are hallucinated from human training samples — StefanoGogioso · 2026-10-01