Event and Entity Coreference Across Five Languages: Effects of Context and Referring Expression

2026-04-19 · Source: Paper Index on ACL Anthology · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Natural Language Processing · Depth: Expert, quick

Summary

A study investigated event and entity coreference across five languages: English, French, German, Italian, and Spanish. It extended prior work on English by examining how different referring expressions corefer with entity and event antecedents, and whether verbal features like argument structure and aspect influence this choice. Using a story-continuation experiment, the research found consistent, non-categorical biases across languages: personal pronouns favored entity coreference, while demonstratives favored event coreference. Antecedent complexity increased the rate of anaphors coreferring with event antecedents, though event completion (aspectual status) did not reach statistical significance despite uniform patterns. The study also included a comparison of referring expressions for entity and event antecedents in a trilingual parallel corpus annotated with coreference, providing a crosslingual view beyond entity-only coreference.

Key takeaway

For research scientists developing natural language processing models, understanding crosslingual coreference biases is crucial. Your models should account for the observed preference of personal pronouns for entity coreference and demonstratives for event coreference across languages. Incorporating antecedent complexity as a feature can improve event coreference resolution, leading to more robust and accurate crosslingual NLP systems.

Key insights

Crosslingual coreference biases show personal pronouns favor entities, demonstratives favor events, influenced by antecedent complexity.

Principles

• Pronoun choice signals coreference type.
• Antecedent complexity impacts event coreference.
• Crosslingual coreference patterns are largely consistent.

Method

A story-continuation experiment tested referring expression preferences for entity and event antecedents across five languages, manipulating verbal features like argument structure and aspect. A trilingual parallel corpus comparison was also conducted.

In practice

• Analyze pronoun usage for coreference disambiguation.
• Consider antecedent complexity in NLP models.
• Apply crosslingual coreference patterns in translation.

Topics

• Event Coreference
• Entity Coreference
• Crosslingual Linguistics
• Referring Expressions
• Personal Pronouns

Best for: Research Scientist, AI Scientist, NLP Engineer

Related on AIssential

• When English Isn't the Best Teacher: Source Language Effects in Cross-Lingual In-Context Learning — ICL cross-lingual transfer differs from fine-tuning, requiring new source language selection heuristics.
• Coreference annotation for the characterization of characters in Portuguese literary works — Coreference annotation enhances character characterization in Portuguese literary corpora for computational literary studies.
• Disparities In Negation Understanding Across Languages In Vision-Language Models — VLMs exhibit affirmation bias, with negation understanding varying significantly across typologically diverse languages.
• Beyond semantics: the challenges of annotating pragmatic and discourse phenomena — Annotating abstract semantic and pragmatic information is challenging due to its implicit, context-dependent nature.
• CorPipe at CRAC 2026: Empty Nodes and Cross-Lingual Transfer in Multilingual Coreference Resolution — CorPipe 26 integrates empty node prediction with mention and coreference resolution for improved multilingual performance.
• Probing Semantic Alignment, Lexical Invariance, and Syntactic Influence in LLM Metaphor Processing — LLMs struggle with deep metaphor comprehension, often misinterpreting concepts and relying on lexical associations over context or full syntactic understanding.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Paper Index on ACL Anthology.