Контрастивті декодтау
method
ID: contrastivedecoding
Related Papers
Contrastive Decoding: Open-ended Text Generation as Optimization
AbstractGiven a language model (LM), maximum probability is a poor decoding objective for open-ended generation, because it produces short and repetitive text. On the other hand, sampling can often produce …...
SCOTT: Self-Consistent Chain-of-Thought Distillation
AbstractLarge language models (LMs) beyond a certain scale, demonstrate the emergent capability of generating free-text rationales for their predictions via chain-of-thought (CoT) prompting. While CoT can yield dramatically improved performance, …...