← sources

Any progress on continuous learning prompt, will we have meaningful weights plasticity or they'll stay forever frozen?

source: reddit (reddit:LocalLLaMA)published 2026-10-31 11:27ingested 2026-11-01 02:40

retained text

So I'm wondering if we til ever have repeated data ingestion without quality degradation and without full retraining. LoRA and friends are excluded, will we stay forever with frozen weights or will be be able to integrate new knowledge here? 

Any promising research?

document fetched 2026-11-01 00:38 · canonical https://reddit.com/r/LocalLLaMA/comments/1xk7srs/any_progress_on_continuous_learning_prompt_will