Tag
parameter-efficient fine-tuning
2 articles
Research/Jun 12
ART fine-tunes multimodal LLMs via pixels
ART fine-tunes frozen multimodal LLMs by optimizing a single input image instead of model weights.
Research/Jun 11
Which LoRA? Multilingual tuning says simpler wins
Massey University finds basic LoRA matches newer variants in multilingual instruction tuning.