2026
LifeAlign: Lifelong Alignment for Large Language Models with Memory-Augmented Focalized Preference Optimization
AAAI 2026technical
Alignment plays a crucial role in Large Language Models (LLMs) in aligning with human preferences on a specific task/domain. Traditional alignment methods suffer from catastrophic forgetting, where models lose previously learned values when adapting to new preferences or domains. We introduce LifeAl