|
|
|
|
|
|
by a_wild_dandan
805 days ago
|
|
|
Yeah, most future applications will use grammar-based sampling. It's trivial now to restrict tokens to valid JSON, schemas, SQL, etc. But we'll need more elaborate grammars for the limitless domains that LLMs will be applied to. A policy of just rawdoggin' any token is...not long for this world. |
|
|