LoRA adapters for new tasks can be generated on a CPU by softmin-weighted mixing of a large bank of pre-trained adapters, using dataset distribution distances as weights.
Table 2 reports the time elapsed at each stage of our LoRA generation pipeline, measured on a Dell XPS15 (Intel i7-13700H, 14 cores, 64 GB RAM)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
LoRA Fine-Tuning Without GPUs: A CPU-Efficient Meta-Generation Framework for LLMs
LoRA adapters for new tasks can be generated on a CPU by softmin-weighted mixing of a large bank of pre-trained adapters, using dataset distribution distances as weights.