Cover Image

The Frequency Confound in Language-Model Surprisal and Metaphor Novelty

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Momen, Omar, Zarrieß, Sina
Format:	Preprint
Published:	2026
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Surprisal and Metaphor Novelty Judgments: Moderate Correlations and Divergent Scaling Effects Revealed by Corpus-Based and Synthetic Datasets
by: Momen, Omar, et al.
Published: (2026)

How Hypocritical Is Your LLM judge? Listener-Speaker Asymmetries in the Pragmatic Competence of Large Language Models
by: Sieker, Judith, et al.
Published: (2026)

Linguistic Structure Induction from Language Models
by: Momen, Omar
Published: (2024)

Dialogue Is Not Enough to Make a Communicative BabyLM (But Neither Is Developmentally Inspired Reinforcement Learning)
by: Padovani, Francesca, et al.
Published: (2025)

Child-directed speech facilitates production, not comprehension, in BabyLMs
by: Bunzeck, Bastian, et al.
Published: (2026)

Subword models struggle with word learning, but surprisal hides it
by: Bunzeck, Bastian, et al.
Published: (2025)

Resilience through Scene Context in Visual Referring Expression Generation
by: Junker, Simeon, et al.
Published: (2024)

SceneGram: Conceptualizing and Describing Tangrams in Scene Context
by: Junker, Simeon, et al.
Published: (2025)

Model Interpretability and Rationale Extraction by Input Mask Optimization
by: Brinner, Marc, et al.
Published: (2025)

Do Construction Distributions Shape Formal Language Learning In German BabyLMs?
by: Bunzeck, Bastian, et al.
Published: (2025)

Are BabyLMs Deaf to Gricean Maxims? A Pragmatic Evaluation of Sample-efficient Language Models
by: Askari, Raha, et al.
Published: (2025)

Rationalizing Transformer Predictions via End-To-End Differentiable Self-Training
by: Brinner, Marc, et al.
Published: (2025)

Efficient Scientific Full Text Classification: The Case of EICAT Impact Assessments
by: Brinner, Marc Felix, et al.
Published: (2025)

Small Language Models Also Work With Small Vocabularies: Probing the Linguistic Abilities of Grapheme- and Phoneme-Based Baby Llamas
by: Bunzeck, Bastian, et al.
Published: (2024)

SemCSE: Semantic Contrastive Sentence Embeddings Using LLM-Generated Summaries For Scientific Abstracts
by: Brinner, Marc, et al.
Published: (2025)

Increasing The Performance of Cognitively Inspired Data-Efficient Language Models via Implicit Structure Building
by: Momen, Omar, et al.
Published: (2023)

LLMs Struggle to Reject False Presuppositions when Misinformation Stakes are High
by: Sieker, Judith, et al.
Published: (2025)

Are Multimodal Large Language Models Pragmatically Competent Listeners in Simple Reference Resolution Tasks?
by: Junker, Simeon, et al.
Published: (2025)

SemCSE-Multi: Multifaceted and Decodable Embeddings for Aspect-Specific and Interpretable Scientific Domain Mapping
by: Brinner, Marc, et al.
Published: (2025)

Reference Games as a Testbed for the Alignment of Model Uncertainty and Clarification Requests
by: Ali, Manar, et al.
Published: (2026)

Enhancing Domain-Specific Encoder Models with LLM-Generated Data: How to Leverage Ontologies, and How to Do Without Them
by: Brinner, Marc, et al.
Published: (2025)

Talking to a Know-It-All GPT or a Second-Guesser Claude? How Repair reveals unreliable Multi-Turn Behavior in LLMs
by: Lachenmaier, Clara, et al.
Published: (2026)

Can LLMs Ground when they (Don't) Know: A Study on Direct and Loaded Political Questions
by: Lachenmaier, Clara, et al.
Published: (2025)

DEPLAIN: A German Parallel Corpus with Intralingual Translations into Plain Language for Sentence and Document Simplification
by: Stodden, Regina, et al.
Published: (2023)

Implicit Causality-biases in humans and LLMs as a tool for benchmarking LLM discourse capabilities
by: Kankowski, Florian, et al.
Published: (2025)

Evaluating Diversity in Automatic Poetry Generation
by: Chen, Yanran, et al.
Published: (2024)

The InviTE Corpus: Annotating Invectives in Tudor English Texts for Computational Modeling
by: Spliethoff, Sophie, et al.
Published: (2025)

Clozing the Gap: Exploring Why Language Model Surprisal Outperforms Cloze Surprisal
by: Nair, Sathvik, et al.
Published: (2026)

NoveltyBench: Evaluating Language Models for Humanlike Diversity
by: Zhang, Yiming, et al.
Published: (2025)

The Illusion of Competence: Evaluating the Effect of Explanations on Users' Mental Models of Visual Question Answering Systems
by: Sieker, Judith, et al.
Published: (2024)

Mitigating Hidden Confounding by Progressive Confounder Imputation via Large Language Models
by: Yang, Hao, et al.
Published: (2025)

Finding Challenging Metaphors that Confuse Pretrained Language Models
by: Li, Yucheng, et al.
Published: (2024)

Multi-Novelty: Improve the Diversity and Novelty of Contents Generated by Large Language Models via inference-time Multi-Views Brainstorming
by: Lagzian, Arash, et al.
Published: (2025)

Mining for Species, Locations, Habitats, and Ecosystems from Scientific Papers in Invasion Biology: A Large-Scale Exploratory Study with Large Language Models
by: D'Souza, Jennifer, et al.
Published: (2025)

The Impact of Token Granularity on the Predictive Power of Language Model Surprisal
by: Oh, Byung-Doh, et al.
Published: (2024)

Language Models are Surprisingly Fragile to Drug Names in Biomedical Benchmarks
by: Gallifant, Jack, et al.
Published: (2024)

Frequency Explains the Inverse Correlation of Large Language Models' Size, Training Data Amount, and Surprisal's Fit to Reading Times
by: Oh, Byung-Doh, et al.
Published: (2024)

Confabulation: The Surprising Value of Large Language Model Hallucinations
by: Sui, Peiqi, et al.
Published: (2024)

Testing the Predictions of Surprisal Theory in 11 Languages
by: Wilcox, Ethan Gotlieb, et al.
Published: (2023)

Harnessing Large Language Models for Scientific Novelty Detection
by: Liu, Yan, et al.
Published: (2025)