Large Language Models
270 views
DPO
Quick Definition
Alignment method bypassing reward model for preference learning
Full Definition
Direct Preference Optimization, a simplified alignment method that bypasses explicit reward model training.
Examples
LLM alignment, preference learning, simpler RLHF alternative
Related Terms
rlhf
alignment
reward-model
More Large Language Models Terms
Prompt Caching
Storing key-value states for repeated prompt prefixes
Pre-training
Initial training learning general language patterns from text
Embeddings LLM
Dense vectors capturing semantic meaning of text
Masked Language Model
Training objective predicting randomly masked input tokens
Instruction Tuning
Fine-tuning on instruction-response pairs for better following
Mistral
Efficient open-source LLMs with strong performance