ULPT optimizes prompts in ultra-low dimensions with frozen random up-projection to cut training parameters by 98% while matching vanilla prompt tuning performance on NLP tasks.
ATTEMPT : Parameter-efficient multi-task tuning via attentional mixtures of soft prompts
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Ultra-Low-Dimensional Prompt Tuning via Random Projection
ULPT optimizes prompts in ultra-low dimensions with frozen random up-projection to cut training parameters by 98% while matching vanilla prompt tuning performance on NLP tasks.