Home  | Publications | ZWH+24

Constructions Are So Difficult That Even Large Language Models Get Them Right for the Wrong Reasons

MCML Authors

Link to Profile Hinrich Schütze PI Matchmaking

Hinrich Schütze

Prof. Dr.

Principal Investigator

Computational Linguistics

Abstract

In this paper, we make a contribution that can be understood from two perspectives: from an NLP perspective, we introduce a small challenge dataset for NLI with large lexical overlap, which minimises the possibility of models discerning entailment solely based on token distinctions, and show that GPT-4 and Llama 2 fail it with strong bias. We then create further challenging sub-tasks in an effort to explain this failure. From a Computational Linguistics perspective, we identify a group of constructions with three classes of adjectives which cannot be distinguished by surface features. This enables us to probe for LLM's understanding of these constructions in various ways, and we find that they fail in a variety of ways to distinguish between them, suggesting that they don't adequately represent their meaning or capture the lexical properties of phrasal heads.

inproceedings

LREC-COLING 2024

Joint International Conference on Computational Linguistics, Language Resources and Evalutaion. Torino, Italy, May 20-25, 2024.

Authors

S. Zhou • L. Weissweiler • T. He • H. Schütze • D. R. Mortensen • L. Levin

Links

URL

Research Area

 B2 | Natural Language Processing

BibTeXKey: ZWH+24

Back to Top