Have We Designed Generalizable Structural Knowledge Promptings? Systematic Evaluation and Rethinking
This work addresses the problem of limited generalization in SKP methods for researchers and practitioners in NLP, but it is incremental as it focuses on evaluation rather than proposing a new method.
The paper systematically evaluates the generalization of Structural Knowledge Prompting (SKP) for improving factual accuracy in large language models, introducing a new benchmark called SUBARU with 9 tasks to assess its capabilities.
Large language models (LLMs) have demonstrated exceptional performance in text generation within current NLP research. However, the lack of factual accuracy is still a dark cloud hanging over the LLM skyscraper. Structural knowledge prompting (SKP) is a prominent paradigm to integrate external knowledge into LLMs by incorporating structural representations, achieving state-of-the-art results in many knowledge-intensive tasks. However, existing methods often focus on specific problems, lacking a comprehensive exploration of the generalization and capability boundaries of SKP. This paper aims to evaluate and rethink the generalization capability of the SKP paradigm from four perspectives including Granularity, Transferability, Scalability, and Universality. To provide a thorough evaluation, we introduce a novel multi-granular, multi-level benchmark called SUBARU, consisting of 9 different tasks with varying levels of granularity and difficulty.