Layer-wise Curriculum Learning for Efficient LLM Compression

A technical method for compressing large language models more efficiently, with no direct safety or governance implications.

Auto-scored: This is a neutral efficiency improvement in model compression, not a risk or mitigation event.