Semantic maps built faster by combining data with expert knowledge
Behave Like Experts: Fusing Data with Semantic Knowledge for Effective and Efficient Semantic Map Construction
Computation and Language
Summary
Building maps that connect ideas based on language is hard because computers usually just rely on data, missing out on expert understanding. The authors show a way to mix data with knowledge stored inside language models to mimic how experts think. This helps the system find better connections faster and more accurately. They tested it on various datasets and it worked well.
What this means in practice
- •For language technology developers: Construct more accurate semantic networks to improve multilingual natural language understanding systems.
- •For machine translation engineers: Speed up semantic model building to enhance concept alignment across languages in translation tools.
Authors
Zhu Liu, Zhen Hu, Yuxuan Liu, Ying Liu
Abstract
Semantic map models (SMMs) construct a graph that reflects conceptual similarity by comparing cross-lingual form-meaning mappings. Experts build such models not only by obeying the data constraint, i.e., the connectivity hypothesis, which states that conceptual nodes belonging to the same form must be connected, but also by incorporating their prior semantic knowledge. Current automatic methods ignore this knowledge, which harms both accuracy and efficiency by enlarging the search space of candidate graphs. In this paper, we fuse the distributional semantic knowledge hidden in the internal representations of language models into a top-down, data-driven algorithm. This expert-like behavior re-sorts the candidates in a huge search space, guided by conceptual semantics similarity. Experiments on a self-collected benchmark of 10 datasets and comprehensive metrics show that our method builds SMMs both effectively and efficiently.