LaMP: When Large Language Models Meet Personalization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Salemi, Alireza, Mysore, Sheshera, Bendersky, Michael, Zamani, Hamed
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914823819755520
author Salemi, Alireza
Mysore, Sheshera
Bendersky, Michael
Zamani, Hamed
author_facet Salemi, Alireza
Mysore, Sheshera
Bendersky, Michael
Zamani, Hamed
contents This paper highlights the importance of personalization in large language models 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. We additionally propose two retrieval augmentation approaches that retrieve personal items from each user profile for personalizing language model outputs. To this aim, we study various retrieval models, 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.
format Preprint
id arxiv_https___arxiv_org_abs_2304_11406
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle LaMP: When Large Language Models Meet Personalization
Salemi, Alireza
Mysore, Sheshera
Bendersky, Michael
Zamani, Hamed
Computation and Language
This paper highlights the importance of personalization in large language models 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. We additionally propose two retrieval augmentation approaches that retrieve personal items from each user profile for personalizing language model outputs. To this aim, we study various retrieval models, 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.
title LaMP: When Large Language Models Meet Personalization
topic Computation and Language
url https://arxiv.org/abs/2304.11406