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

• URL: http://arxiv.org/abs/2604.13713v1
• Date: Wed, 15 Apr 2026 10:48:09 GMT
• Title: Learning the Cue or Learning the Word? Analyzing Generalization in Metaphor Detection for Verbs
• Abstract summary: We analyze metaphor detection through RoBERTa, the shared backbone of many state-of-the-art systems, focusing on English verbs.<n>We introduce a controlled lexical hold-out setup where all instances of selected target lemmas are strictly excluded from fine-tuning, and compare predictions on these Held-out lemmas against Exposed lemmas.<n>While the model performs best on Exposed lemmas, it maintains robust performance on Held-out lemmas.
• Score: 54.86138668387175
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Metaphor detection models achieve strong benchmark performance, yet it remains unclear whether this reflects transferable generalization or lexical memorization. To address this, we analyze generalization in metaphor detection through RoBERTa, the shared backbone of many state-of-the-art systems, focusing on English verbs using the VU Amsterdam Metaphor Corpus. We introduce a controlled lexical hold-out setup where all instances of selected target lemmas are strictly excluded from fine-tuning, and compare predictions on these Held-out lemmas against Exposed lemmas (verbs seen during fine-tuning). While the model performs best on Exposed lemmas, it maintains robust performance on Held-out lemmas. 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. Together, these results suggest that generalization is primarily driven by "learning the cue" (transferable contextual patterns), while "learning the word" (verb-specific memorization) provides an additive boost when lexical exposure is available.

Related papers

• Floating or Suggesting Ideas? A Large-Scale Contrastive Analysis of Metaphorical and Literal Verb-Object Constructions [53.690096725532726]
  We analyze 297 English verb-object pairs (e.g., float idea vs. suggest idea) in 2M corpus sentences.<n>Using five NLP tools, we extract 2,293 cognitive and linguistic features capturing affective, lexical, syntactic, and discourse-level properties.<n>Cross-pair results show literal contexts have higher lexical frequency, cohesion, and structural regularity, while metaphorical contexts show greater affective load, imageability, lexical diversity, and constructional specificity.
  arXiv  Detail & Related papers  (2026-04-09T14:08:57Z)
• LANE: Lexical Adversarial Negative Examples for Word Sense Disambiguation [3.506940838682547]
  Fine-grained word meaning resolution remains a critical challenge for neural language models.<n>We propose a novel adversarial training strategy, called LANE, to address this limitation.
  arXiv  Detail & Related papers  (2025-11-14T12:37:20Z)
• Enhancing Metaphor Detection through Soft Labels and Target Word Prediction [3.7676096626244986]
  We develop a prompt learning framework specifically designed for metaphor detection. We also introduce a teacher model to generate valuable soft labels. Experimental results demonstrate that our model has achieved state-of-the-art performance.
  arXiv  Detail & Related papers  (2024-03-27T04:51:42Z)
• Transition-based Abstract Meaning Representation Parsing with Contextual Embeddings [0.0]
  We study a way of combing two of the most successful routes to meaning of language--statistical language models and symbolic semantics formalisms--in the task of semantic parsing. We explore the utility of incorporating pretrained context-aware word embeddings--such as BERT and RoBERTa--in the problem of parsing.
  arXiv  Detail & Related papers  (2022-06-13T15:05:24Z)
• Does BERT really agree ? Fine-grained Analysis of Lexical Dependence on a Syntactic Task [70.29624135819884]
  We study the extent to which BERT is able to perform lexically-independent subject-verb number agreement (NA) on targeted syntactic templates. Our results on nonce sentences suggest that the model generalizes well for simple templates, but fails to perform lexically-independent syntactic generalization when as little as one attractor is present.
  arXiv  Detail & Related papers  (2022-04-14T11:33:15Z)
• Did the Cat Drink the Coffee? Challenging Transformers with Generalized Event Knowledge [59.22170796793179]
  Transformers Language Models (TLMs) were tested on a benchmark for the textitdynamic estimation of thematic fit Our results show that TLMs can reach performances that are comparable to those achieved by SDM. However, additional analysis consistently suggests that TLMs do not capture important aspects of event knowledge.
  arXiv  Detail & Related papers  (2021-07-22T20:52:26Z)
• Infusing Finetuning with Semantic Dependencies [62.37697048781823]
  We show that, unlike syntax, semantics is not brought to the surface by today's pretrained models. We then use convolutional graph encoders to explicitly incorporate semantic parses into task-specific finetuning.
  arXiv  Detail & Related papers  (2020-12-10T01:27:24Z)
• On Vocabulary Reliance in Scene Text Recognition [79.21737876442253]
  Methods perform well on images with words within vocabulary but generalize poorly to images with words outside vocabulary. We call this phenomenon "vocabulary reliance" We propose a simple yet effective mutual learning strategy to allow models of two families to learn collaboratively.
  arXiv  Detail & Related papers  (2020-05-08T11:16:58Z)
• Lexical Semantic Recognition [15.944476972081345]
  We train a neural CRF sequence tagger and evaluate its performance along various axes of annotation. Our work also establishes baseline models and evaluation metrics for integrated and accurate modeling of lexical semantics.
  arXiv  Detail & Related papers  (2020-04-30T17:52:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.