Revisiting a Pain in the Neck: A Semantic Reasoning Benchmark for Language Models

TL;DR

SemanticQA is a comprehensive benchmark for evaluating language models on semantic phrase processing, revealing significant performance differences and highlighting areas for improvement in semantic reasoning.

Contribution

It consolidates diverse resources into a unified testbed for semantic reasoning, providing new insights into language models' understanding of complex semantic expressions.

Findings

Performance varies significantly across tasks requiring semantic reasoning.

Language models show differing capabilities in extraction, classification, and interpretation of semantic phrases.

SemanticQA reveals gaps in models' semantic understanding, guiding future improvements.

Abstract

We present SemanticQA, an evaluation suite designed to assess language models (LMs) in semantic phrase processing tasks. The benchmark consolidates existing multiword expression (MwE) resources and reorganizes them into a unified testbed. It covers both general lexical phenomena, such as lexical collocations, and three fine-grained categories: idiomatic expressions, noun compounds, and verbal constructions. Through SemanticQA, we assess LMs of diverse architectures and scales in extraction, classification, and interpretation tasks, as well as sequential task compositions. We reveal substantial performance variation, particularly on tasks requiring semantic reasoning, highlighting differences in reasoning efficacy and semantic understanding of LMs, providing insights for pushing LMs with stronger comprehension on non-trivial semantic phrases. The evaluation harness and data of SemanticQA are available at https://github.com/jacklanda/SemanticQA.