Coreference annotation for the characterization of characters in Portuguese literary works

2026-04-12 · Source: Paper Index on ACL Anthology · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Expert, short

Summary

Diana Santos, Luisa Mara Silva Lima, and Emanoel Pires describe the integration of coreference annotation into public literary corpora to facilitate character characterization in distant reading, a technique used in computational literary studies. Presented at the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) in Salvador, Brazil, this work details the motivation for this task and its implementation using BRAT to ensure legibility and revisability for literary scholars. The authors present initial results from this annotation effort, introduce a small public annotated corpus, and discuss the development of two distinct coreference modules. This research, published by the Association for Computational Linguistics, spans pages 572–580 of the proceedings.

Key takeaway

For NLP Engineers working on literary analysis or digital humanities projects, integrating coreference annotation can significantly improve the depth of character characterization. Your efforts to make annotation legible and revisable, perhaps by using tools like BRAT, will foster better collaboration with literary scholars and enhance the utility of your annotated corpora for distant reading applications.

Key insights

Coreference annotation enhances character characterization in Portuguese literary corpora for computational literary studies.

Principles

• Interdisciplinary tools improve literary analysis.
• Public corpora benefit from detailed linguistic annotation.

Method

Coreference is added to literary corpora, made legible via BRAT, and supported by two coreference modules for character characterization in distant reading.

In practice

• Use BRAT for revisable linguistic annotation.
• Develop specific modules for coreference tasks.

Topics

• Coreference Annotation
• Character Characterization
• Literary Corpora
• Distant Reading
• Computational Literary Studies

Best for: AI Scientist, NLP Engineer, Research Scientist

Related on AIssential

• A Larger Annotated Corpus of Portuguese Coreference — LLMs can annotate Portuguese coreference, but a new LLM-based system did not outperform existing rule-based methods.
• Coreference Resolution in spaCy — spaCy's experimental "coref" component enables robust coreference resolution for NLP applications.
• Identifying Signs and Symptoms of Urinary Tract Infection from Emergency Department Clinical Notes Using Large Language Models — Prodigy is an efficient, scriptable annotation tool enabling data scientists to accelerate NLP development.
• TantraTagger: A Benchmark Dataset for Tantrayukti-Based Discourse Structure Labelling in Sanskrit Śāstra Texts — TantraTagger is a benchmark dataset for discourse structure labelling in Sanskrit Śāstra Texts.
• Characterizing Narrative Content in Web-scale LLM Pretraining Data — Narrative structure in web-scale LLM pretraining data is measurable and unevenly distributed, impacting model reasoning.
• STPar: A Structure-Aware Triaffine Parser for Screenplay Character Coreference Resolution — STPar enhances movie screenplay coreference resolution by integrating structural, triaffine, and multi-task learning approaches.

Open in AIssential →

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