4 ms·
Telling the LLM that it must do something is not a guarantee that it'll follow through.
by computerex 3y ago
Telling the LLM that it must do something is not a guarantee that it'll follow through.
- vykthur 3y agoTrue. This is an open area of research. Tools like guidance (or other implementations of constrained decoding with llms [1,2]) will likely help improve this problem. [1] A guidance language for controlling large language models. https://github.com/guidance-ai/guidance https://github.com/guidance-ai/guidance [2] Knowledge Infused Decoding https://arxiv.org/abs/2204.03084 https://arxiv.org/abs/2204.03084