kogasa, 1 year ago Semantic embeddings are a thing. LLMs “work with tokens” but they associate them with semantic models internally. You can externalize it via semantic embeddings so that the same semantic models can be shared between LLMs.
Semantic embeddings are a thing. LLMs “work with tokens” but they associate them with semantic models internally. You can externalize it via semantic embeddings so that the same semantic models can be shared between LLMs.