Token Prepending: A Training-Free Approach for Eliciting Better Sentence Embeddings from LLMs

LLMAG
Re-assign community