SPOR: A Comprehensive and Practical Evaluation Method for Compositional Generalization in Data-to-Text Generation
This work addresses the need for more practical and comprehensive evaluation frameworks for compositional generalization in data-to-text generation, though it is incremental in extending existing evaluation concepts.
The authors tackled the limited evaluation of compositional generalization in data-to-text generation by proposing SPOR, a comprehensive method covering four manifestations, and found that existing models, including LLMs, are deficient in these aspects.
Compositional generalization is an important ability of language models and has many different manifestations. For data-to-text generation, previous research on this ability is limited to a single manifestation called Systematicity and lacks consideration of large language models (LLMs), which cannot fully cover practical application scenarios. In this work, we propose SPOR, a comprehensive and practical evaluation method for compositional generalization in data-to-text generation. SPOR includes four aspects of manifestations (Systematicity, Productivity, Order invariance, and Rule learnability) and allows high-quality evaluation without additional manual annotations based on existing datasets. We demonstrate SPOR on two different datasets and evaluate some existing language models including LLMs. We find that the models are deficient in various aspects of the evaluation and need further improvement. Our work shows the necessity for comprehensive research on different manifestations of compositional generalization in data-to-text generation and provides a framework for evaluation.