When Thoughts Become Memory: Unifying In-Context Learning and Fine-Tuning
arXiv:2610.29419v1 Announce Type: new Abstract: We are the first to give a certified rule for whether unseen data belong in a pretrained model's context or its weights, built on an exact law for their information value. That value, the code length saved on held-out queries, equals in expectation th…
Read the full story at arXiv stat.ML ↗
Timeline · 1 report
- 2026-10-22 04:00 · arXiv stat.ML
When Thoughts Become Memory: Unifying In-Context Learning and Fine-Tuning