> LLMs do just interpolate their training data "interpolate" has a technical mea... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

jmmcd 4 months ago | parent | context | favorite | on: An OpenAI model has disproved a central conjecture...

> LLMs do just interpolate their training data

"interpolate" has a technical meaning - in this meaning, LLMs almost never interpolate. It also has a very vague everyday meaning - in this meaning, LLMs do interpolate, but so do humans.

astrange 4 months ago [–]

An LLM in a harness with any tools (even a calculator) doesn't just interpolate because it can reach states out of its own distribution.

Consider applying for YC's Winter 2027 batch! Applications are open till November 2.
Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact