Taha Ceritli

model merging (3) large language model (2) parameter-efficient fine-tuning (2) low-rank adaptation (2) parameter efficient fine-tuning (1) parameter efficiency (1) parameter generation (1) language model fine-tuning (1) performance degradation (1) decentralized learning (1) low-rank adapter (1) adapter merging (1) on-device deployment (1) storage efficiency (1) hybrid datum (1) model compression (1) model atla (1) federated learning (1) continual learning (1) variance reduction (1)