From: Can Temporary Context Become Stable Parameters?

When designing an AI memory system, which facts should remain in retrieval context and which deserve parameter updates?

The choice determines reversibility, auditability, latency, and contamination risk. Treating every useful context as a weight update can make correction and provenance harder.

Action

Create a policy classifying information by persistence, sensitivity, update frequency, and required audit trail.

Read the full exploration

What else is in this exploration

3 evidence blocks3 perspectives2 visualizations8 media resources4 rabbit holes

evidence

Large-language-model parameters are static and read-only during inference.

evidence

For standard transformer architectures, the effect of in-context learning cannot be converted exa...

evidence

In-context learning can adapt a model's output during inference without explicit gradient updates...

Continue exploring

Can Temporary Context Become Stable Parameters?

Evidence, perspectives, rabbit holes, and more