作者位次 #2/3ACL 2026Poster· Natural Language Processing
Investigating More Explainable and Partition-Free Compositionality Estimation for LLM s: A Rule-Generation Perspective
Ziyao Xu · Cong Wang · Houfeng Wang
摘要
Compositional generalization tests are often used to estimate the compositionality of LLMs. However, such tests have the following limitations: (1) they only focus on the output results without considering LLMs’ understanding of sample compositionality, resulting in explainability defects; (2) th...