Prompt tuning matches full model tuning performance on large language models while tuning only a small fraction of parameters and improves robustness to domain shifts.
Khapra, and Karthik Sankaranarayanan
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.CL 3representative citing papers
SHINE trains a scalable in-context hypernetwork to generate high-quality LoRA adapters from contexts in one pass, enabling efficient LLM adaptation that saves time and compute compared to standard fine-tuning.
UltraChat supplies 1.5 million high-quality multi-turn dialogues that, when used to fine-tune LLaMA, produce UltraLLaMA, which outperforms prior open-source chat models including Vicuna.
citing papers explorer
-
The Power of Scale for Parameter-Efficient Prompt Tuning
Prompt tuning matches full model tuning performance on large language models while tuning only a small fraction of parameters and improves robustness to domain shifts.
-
SHINE: A Scalable In-Context Hypernetwork for Mapping Context to LoRA in a Single Pass
SHINE trains a scalable in-context hypernetwork to generate high-quality LoRA adapters from contexts in one pass, enabling efficient LLM adaptation that saves time and compute compared to standard fine-tuning.
-
Enhancing Chat Language Models by Scaling High-quality Instructional Conversations
UltraChat supplies 1.5 million high-quality multi-turn dialogues that, when used to fine-tune LLaMA, produce UltraLLaMA, which outperforms prior open-source chat models including Vicuna.