ASPIRE
A framework to enhance the selective prediction capability of large language models
CommonProductProductivityArtificial IntelligenceNatural Language Processing
ASPIRE is a well-designed framework designed to boost the selective prediction capability of large language models. It leverages parameter-efficient fine-tuning training to enable LLMs to self-assess and provide confidence scores for generated answers. Experimental results indicate that ASPIRE significantly outperforms current selective prediction methods on various question-answering datasets.
ASPIRE Visit Over Time
Monthly Visits
4663
Bounce Rate
92.31%
Page per Visit
1.1
Visit Duration
00:00:32