Learning the Cue or Learning the Word? Analyzing Generalization in Metaphor Detection for Verbs

MCML Authors

Sinan Kurtyigit

→ Group Alexander Fraser
 Data Analytics & Statistics

Alexander Fraser

Prof. Dr.

Core PI

Data Analytics & Statistics

Abstract

We analyze the generalization capabilities of a RoBERTa-based metaphor detection model on English verbs using the VU Amsterdam Metaphor Corpus. Since RoBERTa encoders are widely-used in metaphor detection, determining whether they rely on lexical memorization or learn transferable patterns is critical. We introduce a controlled lexical hold-out setup where all instances of selected target lemmas are strictly excluded from fine-tuning. We then compare model predictions on these Held-out lemmas against Exposed lemmas (verbs encountered during fine-tuning). While the model performs best on Exposed lemmas, it maintains robust performance on Held-out lemmas. This success does not correlate with verb lemma frequency, suggesting the generalization is likely not driven by frequency bias in pre-training. Further analysis reveals that sentence context alone is sufficient to match full-model performance on Held-out lemmas, whereas static verb-level embeddings are not. Geometrically, we find that the encoder projects Held-out instances into the same task-relevant subspaces as Exposed instances. Together, these results suggest that the model exhibits generalization capabilities, which appear to be driven by 'learning the cue' (transferable contextual patterns) rather than 'learning the word (verb-specific memorization).

inproceedings KSF+26

---

EMNLP 2026

Conference on Empirical Methods in Natural Language Processing. Budapest, Hungary, Oct 24-29, 2026. To be published. Preprint available.

---

Authors

S. Kurtyigit • S. Schulte im Walde • A. Fraser

---

Links

URL

---

Research Area

 B2 | Natural Language Processing