Synthetic Data Generation for Training Diversified Commonsense Reasoning Models
This addresses the problem of training conversational agents to produce diverse and high-quality commonsense responses, though it is incremental as it builds on existing GCR methods by introducing synthetic data generation.
The paper tackles the lack of large-scale diverse training datasets for generative commonsense reasoning by proposing a two-stage method to create the synthetic dataset CommonSyn, which when used for fine-tuning increases both generation diversity and quality across different LLMs compared to vanilla models and human-crafted datasets.
Conversational agents are required to respond to their users not only with high quality (i.e. commonsense bearing) responses, but also considering multiple plausible alternative scenarios, reflecting the diversity in their responses. Despite the growing need to train diverse commonsense generators, the progress of this line of work has been significantly hindered by the lack of large-scale high-quality diverse commonsense training datasets. Due to the high annotation costs, existing Generative Commonsense Reasoning (GCR) datasets are created using a small number of human annotators, covering only a narrow set of commonsense scenarios. To address this training resource gap, we propose a two-stage method to create the first-ever synthetic dataset CommonSyn for diversified (GCR). The model fine-tuned on our synthetic data jointly increase both generation diversity and quality compared with vanilla models and the model fine-tuned on human-crafted dataset across different size Large Language Models (LLMs)