[Lions: 1] and [Tigers: 2] and [Bears: 3], Oh My! Literary Coreference Annotation with LLMs

2024/01/31 by Rebecca M. M. Hicke, David Mimno, Hicke, Rebecca M. M. +1 · 4 citations
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Natural Language Processing Techniques #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2401.17922

openalex publication_date 2024/01/31 · openalex created_date 2024/02/02 · openalex updated_date 2026/07/28

Abstract

Coreference annotation and resolution is a vital component of computational literary studies. However, it has previously been difficult to build high quality systems for fiction. Coreference requires complicated structured outputs, and literary text involves subtle inferences and highly varied language. New language-model-based seq2seq systems present the opportunity to solve both these problems by learning to directly generate a copy of an input sentence with markdown-like annotations. We create, evaluate, and release several trained models for coreference, as well as a workflow for training new models.

Cited by

• The Elephant in the Coreference Room: Resolving Coreference in Full-Length French Fiction Works
• Findings of the Fourth Shared Task on Multilingual Coreference Resolution: Can LLMs Dethrone Traditional Approaches?
• Referential ambiguity and clarification requests: comparing human and LLM behaviour
• Evaluating Prompt-Based and Fine-Tuned Approaches to Czech Anaphora Resolution

Related

• The Elephant in the Coreference Room: Resolving Coreference in Full-Length French Fiction Works
• Coreference Resolution for Full German Novels using Large Language Models
• Are Large Language Models Robust Coreference Resolvers?
• Model-based annotation of coreference
• CoreLM: Coreference-aware Language Model Fine-Tuning