Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Synthetic Data Generation with Large Language Models for Personalized Community Question Answering

Marco Braga, Pranav Kasela, Alessandro Raganato, Gabriella Pasi

TL;DR

The potential of Large Language Models for generating synthetic documents to train an IR system for a Personalized Community Question Answering task and the introduction of a new dataset, named Sy-SE-PQA, suggest that LLMs have high potential in generating data tailored to users' needs.

Abstract

Personalization in Information Retrieval (IR) is a topic studied by the research community since a long time. However, there is still a lack of datasets to conduct large-scale evaluations of personalized IR; this is mainly due to the fact that collecting and curating high-quality user-related information requires significant costs and time investment. Furthermore, the creation of datasets for Personalized IR (PIR) tasks is affected by both privacy concerns and the need for accurate user-related data, which are often not publicly available. Recently, researchers have started to explore the use of Large Language Models (LLMs) to generate synthetic datasets, which is a possible solution to generate data for low-resource tasks. In this paper, we investigate the potential of Large Language Models (LLMs) for generating synthetic documents to train an IR system for a Personalized Community Question Answering task. To study the effectiveness of IR models fine-tuned on LLM-generated data, we introduce a new dataset, named Sy-SE-PQA. We build Sy-SE-PQA based on an existing dataset, SE-PQA, which consists of questions and answers posted on the popular StackExchange communities. Starting from questions in SE-PQA, we generate synthetic answers using different prompt techniques and LLMs. Our findings suggest that LLMs have high potential in generating data tailored to users' needs. The synthetic data can replace human-written training data, even if the generated data may contain incorrect information.

Synthetic Data Generation with Large Language Models for Personalized Community Question Answering

TL;DR

The potential of Large Language Models for generating synthetic documents to train an IR system for a Personalized Community Question Answering task and the introduction of a new dataset, named Sy-SE-PQA, suggest that LLMs have high potential in generating data tailored to users' needs.

Abstract

Personalization in Information Retrieval (IR) is a topic studied by the research community since a long time. However, there is still a lack of datasets to conduct large-scale evaluations of personalized IR; this is mainly due to the fact that collecting and curating high-quality user-related information requires significant costs and time investment. Furthermore, the creation of datasets for Personalized IR (PIR) tasks is affected by both privacy concerns and the need for accurate user-related data, which are often not publicly available. Recently, researchers have started to explore the use of Large Language Models (LLMs) to generate synthetic datasets, which is a possible solution to generate data for low-resource tasks. In this paper, we investigate the potential of Large Language Models (LLMs) for generating synthetic documents to train an IR system for a Personalized Community Question Answering task. To study the effectiveness of IR models fine-tuned on LLM-generated data, we introduce a new dataset, named Sy-SE-PQA. We build Sy-SE-PQA based on an existing dataset, SE-PQA, which consists of questions and answers posted on the popular StackExchange communities. Starting from questions in SE-PQA, we generate synthetic answers using different prompt techniques and LLMs. Our findings suggest that LLMs have high potential in generating data tailored to users' needs. The synthetic data can replace human-written training data, even if the generated data may contain incorrect information.

Paper Structure

This paper contains 19 sections, 2 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Large Language Models
  4. Synthetic Data Generation
  5. Methodology
  6. Dataset Generation
  7. Basic Answer Generation
  8. Personalized Answer Generation
  9. Contextual Answer Generation
  10. Baselines and Evaluation
  11. Results
  12. Exploratory Analysis
  13. Data Diversity
  14. Factual Information in generated answers
  15. Conclusion and Future Works
  16. ...and 4 more sections

Figures (2)

  • Figure 1: Illustration of the process of Data Generation and Evaluation. In the Data Generation step, we generate the synthetic answers to questions sampled from StackExchange. Then we train neural retrieval models where the queries are the questions and the documents the synthetic answers. Finally, in the Evaluation step, we evaluate the retrieval models on a human-written test set.
  • Figure 2: Example of a question with the three prompts utilized.