AI RESEARCH

Aletheia: Gradient-Guided Layer Selection for Efficient LoRA Fine-Tuning Across Architectures

arXiv CS.LG

ArXi:2604.15351v1 Announce Type: new Low-Rank Adaptation (LoRA) has become the dominant parameter-efficient fine-tuning method for large language models, yet standard practice applies LoRA adapters uniformly to all transformer layers regardless of their relevance to the downstream task. We