Latent schema reasoning improves question answering over knowledge bases
Schema-Anchored Latent Reasoning for Semantic Parsing-Based Knowledge Base Question Answering
Computation and Language
Summary
Answering complex questions using large knowledge bases requires understanding which parts of the data to look at and how to combine them. The authors introduce a method called SALR that lets a language model think through its reasoning steps in a hidden, flexible way, delaying final decisions until more information is considered. SALR links these hidden thoughts to parts of the knowledge base to help guide the final answer without forcing explicit step-by-step explanations. Their tests show this approach helps the model produce more accurate answers, especially for complicated questions.
What this means in practice
- •For software developers: Improve question answering systems by integrating latent schema alignment that enhances reasoning over large, complex knowledge bases.
- •For digital assistants teams: Enhance the ability of voice and chat assistants to accurately interpret and answer multi-step questions using complex databases.
Authors
Guangze Gao, Zixuan Li, Sikui Zhang, Chunfeng Yuan, Wenjuan Li, Bing Li, Xiaolong Jin, Weiming Hu
Abstract
Semantic parsing (SP)-based knowledge base question answering aims to answer natural language questions by generating executable logical forms (LFs) over knowledge bases (KBs). When applying Large Language Models (LLMs) to this task, a key challenge over large, heterogeneous KBs is selecting question-related schema elements (i.e., relations and classes) and composing them into complex LFs. Recent LLM-based methods often make early discrete commitments to schema elements during intermediate reasoning, allowing incorrect intermediate schema decisions to propagate and finally result in incorrect LFs. To overcome this limitation, we propose SALR, a schema-anchored latent reasoning method for LF construction. It performs multi-step reasoning by generating continuous thoughts in the model's hidden states, thereby delaying the explicit commitment to LF decisions. To ground this latent reasoning process in the corresponding KB schema, SALR aligns continuous thoughts with a codebook of KB schema elements through an alignment objective supervised by schema traces deterministically derived from gold LFs. It then incorporates the aligned schema codes into inputs for subsequent reasoning steps. This schema-mediated feedback guides LF generation without requiring the model to emit an explicit textual reasoning trajectory. Experiments on GrailQA and WebQSP show that SALR achieves consistent overall gains over strong baselines. Notably, on compositional questions from GrailQA, SALR outperforms TIARA, a strong SP-based baseline, by 2.86 F1 points. Further analyses show that schema-mediated feedback affects LF generation and that schema information is recoverable from the latent states.