Investigating the Impact of Conceptual Metaphors on LLM-Based NLI Through Shapley Interactions

MCML Authors

Maximilian Muschalik

→ Group Eyke Hüllermeier
 Artificial Intelligence and Machine Learning

Fabian Fumagalli

Prof. Dr.

Thomas Bayes Fellow

* Former Thomas Bayes Fellow

Eyke Hüllermeier

Prof. Dr.

Core PI

Artificial Intelligence and Machine Learning

Abstract

Metaphorical language is prevalent in everyday communication, often used unconsciously, as in “rising crime.” While LLMs excel at identifying metaphors in text, they struggle with downstream tasks that implicitly require correct metaphor interpretation, such as natural language inference (NLI). This work explores how LLMs perform on NLI with metaphorical input. Particularly, we investigate whether incorporating conceptual metaphors (source and target domains) enhances performance in zero-shot and few-shot settings. Our contributions are two-fold: (1) we extend metaphorical texts in an existing NLI dataset by source and target domains, and (2) we conduct an ablation study using Shapley values and interactions to assess the extent to which LLMs interpret metaphorical language correctly in NLI. Our results indicate that incorporating conceptual metaphors often improves task performance.

inproceedings SMF+25

---

Findings @EMNLP 2025

Findings of the Conference on Empirical Methods in Natural Language Processing. Suzhou, China, Nov 04-09, 2025.

---

Authors

M. Sengupta • M. Muschalik • F. Fumagalli • B. Hammer • E. Hüllermeier • D. Ghosh • H. Wachsmut

---

Links

DOI

---

In Collaboration

 Analog Devices

---

Research Areas

 A1 | Statistical Foundations & Explainability

 A3 | Computational Models