LaMP: Personalization Benchmark for LLMs
This paper highlights the importance of personalization in LLMs and introduces the LaMP benchmark — a novel benchmark for training and evaluating language models for producing personalized outputs. LaMP offers a comprehensive evaluation framework with diverse language tasks and multiple entries for each user profile. It consists of seven personalized tasks, spanning three text classification and four text generation tasks. Two retrieval augmentation approaches are proposed that retrieve personal items from each user profile for personalizing language model outputs. To this aim, various retrieval models are studied, including term matching, semantic matching, and time-aware methods. Extensive experiments on LaMP for zero-shot and fine-tuned language models demonstrate the efficacy of the proposed retrieval augmentation approach and highlight the impact of personalization in various natural language tasks.
In this video, I talk about the following: What is the LaMP benchmark? How does FLAN-T5 base perform on LaMP? How do LLMs perform on LaMP (zero-shot)?
For more details, please look at https://arxiv.org/pdf/2304.11406
Salemi, Alireza, Sheshera Mysore, Michael Bendersky, and Hamed Zamani. "LaMP: When large language models meet personalization." arXiv preprint arXiv:2304.11406 (2023).