|
|
|
|
|
|
by rckrd
1134 days ago
|
|
|
For a less dramatic strategy with LLMs that expose the tokenizer vocabulary, you can use context-free grammars to constrain the logits according to the parser so that the LLMs only generate valid next tokens for the language.[0] [0]https://github.com/r2d4/parserllm |
|
|