Large Language Models
269 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
Mixture of Experts
Architecture routing tokens to specialized expert sub-networks
RLHF
Aligning LLMs with human preferences using reinforcement learning
Fine-Tuning
Further training pre-trained models on specific domain data
Jailbreaking
Circumventing AI safety restrictions through prompt techniques
Positional Encoding
Incorporating token position information into transformers
Training LLM
Adjusting parameters on large datasets to learn language