Search papers, labs, and topics across Lattice.
This paper introduces SymCA, an interpretable column annotation framework that leverages large language models (LLMs) to enhance the accuracy and interpretability of column type and property annotations. By employing a two-component approach鈥攇lobal skeleton induction for semantic structure and local substrate evolution for predictive modeling鈥擲ymCA effectively captures rich label semantics while maintaining a transparent decision-making process. Experimental results show that SymCA significantly outperforms existing methods, achieving an average improvement of 6.42% in Micro-F1 and 11.03% in Macro-F1 scores.
SymCA achieves a remarkable balance between accuracy and interpretability in column annotation, outperforming traditional methods by over 6% in F1 scores.
Column annotation (CA), including column type annotation (CTA) and column property annotation (CPA), aims to identify the meanings of table columns and the semantic relationships among them. Recent CA methods usually use various neural models to learn column representations and directly map them to label categories, thereby (1) sacrificing model interpretability and adaptivity, and (2) overlooking rich label semantics and ultimately limiting accuracy. To address these limitations, we propose SymCA, an LLM-empowered interpretable CA framework that materializes column annotation as a global-to-local symbolic decision process. SymCA consists of two components: (1) global skeleton induction, which constructs a semantic skeleton over the label space, and (2) local substrate evolution, which evolves predictive substrates within the skeleton. Specifically, to exploit label semantics while preserving an interpretable decision process, the global skeleton induction module leverages LLMs to generate candidate hypernym-inspired tree-structured semantic skeletons and employs a Minimum Bayes Risk (MBR)-based consensus strategy to select a robust skeleton against generation variance. Since different internal nodes require different evidence to distinguish among their child nodes, the local substrate evolution module materializes each internal node as an executable and evolvable predictive substrate. Over multiple evolution rounds, each substrate trains an interpretable random forest classifier with the current operator set, leverages the LLM to propose node-specific operator modifications, and uses an exploration-exploitation strategy to prioritize promising substrates. Extensive experiments demonstrate that SymCA is accurate, robust, and interpretable, outperforming the strongest baselines by an average of 6.42% in Micro-F1 and 11.03% in Macro-F1.
