Reasoning about Ambiguous Definite Descriptions

Stefan Frederik Schouten, Peter Bloem, Ilia Markov, Piek Vossen

Submission Type: Regular Short Paper

Submission Track: Language Modeling and Analysis of Language Models

Submission Track 2: Resources and Evaluation

Keywords: Natural language reasoning, definite descriptions, ambiguity, large language models

Abstract:

Natural language reasoning plays an increasingly important role in improving language models' ability to solve complex language understanding tasks. An interesting use case for reasoning is the resolution of context-dependent ambiguity. But no resources exist to evaluate how well Large Language Models can use explicit reasoning to resolve ambiguity in language. We propose to use ambiguous definite descriptions for this purpose and create and publish the first benchmark dataset consisting of such phrases. Our method includes all information required to resolve the ambiguity in the prompt, which means a model does not require anything but reasoning to do well.
We find this to be a challenging task for recent LLMs. Code and data available at: https://github.com/sfschouten/exploiting-ambiguity

Submission Number: 4256