Select, Prompt, Filter: Distilling Large Language Models for Summarizing Conversations
Submission Type: Regular Short Paper
Submission Track: Summarization
Submission Track 2: Theme Track: Large Language Models and the Future of NLP
Keywords: Text Summarization, Generative AI, Knowledge distillation, Data filtering
TL;DR: Data synthesis for summarizing conversations
Abstract: Large language models (LLMs) like ChatGPT can be expensive to train, deploy, and use for specific natural language generation tasks such as text summarization and for certain domains. A promising alternative is to fine-tune relatively smaller language models (LMs) on a particular task using high-quality, in-domain datasets. However, it can be prohibitively expensive to get such high-quality training data. This issue has been mitigated by generating weakly supervised data via knowledge distillation (KD) of LLMs. We propose a three-step approach to distill ChatGPT and fine-tune smaller LMs for summarizing forum conversations. More specifically, we design a method to selectively sample a large unannotated corpus of forum conversation using a semantic similarity metric. Then, we use the same metric to retrieve suitable prompts for ChatGPT from a small annotated validation set in the same domain. The generated dataset is then filtered to remove low-quality instances. Our proposed select-prompt-filter KD approach leads to significant improvements of up to 6.6 ROUGE-2 score by leveraging sufficient in-domain pseudo-labeled data over a standard KD approach given the same size of training data.
Submission Number: 2824
Loading