An Investigation of LLMs’ Inefficacy in Understanding Converse Relations

Chengwen Qi, Bowen Li, Binyuan Hui, Bailin Wang, Jinyang Li, Jinwang Wu, Yuanjun Laili

Submission Type: Regular Long Paper

Submission Track: Language Modeling and Analysis of Language Models

Submission Track 2: Interpretability, Interactivity, and Analysis of Models for NLP

Keywords: Large language models, inverse scaling, converse relation

TL;DR: We show that the larger LLMs are, the harder they fail on understanding converse relations.

Abstract:

Large Language Models (LLMs) have achieved remarkable success in many formal language oriented tasks, such as structural data-to-text and semantic parsing.
However current benchmarks mostly follow the data distribution of the pre-training data of LLMs.
Therefore, a natural question rises that do LLMs really understand the structured semantics of formal languages.
In this paper, we investigate this problem on a special case, converse binary relation.
We introduce a new benchmark ConvRe focusing on converse relations, which contains 17 relations and 1240 triples extracted from popular knowledge graph completion datasets.
Our ConvRE features two tasks, Re2Text and Text2Re, which are formulated as multi-choice question answering to evaluate LLMs' ability to determine the matching between relations and associated text.
For the evaluation protocol, apart from different prompting methods, we further introduce variants to the test text and few-shot example text.
We conduct experiments on three popular LLM families and have observed various scaling trends.
The results suggest that LLMs
often resort to shortcut learning and still face challenges on our proposed benchmark.

Submission Number: 876