Cross-lingual Cross-temporal Summarization: Dataset, Models, Evaluation

TL;DR

This paper introduces the first dataset and evaluation framework for cross-lingual cross-temporal summarization, comparing transformer models and GPT-3.5, revealing GPT-3.5's relative effectiveness and challenges in summarizing historical texts across languages.

Contribution

It creates a novel CLCTS dataset, evaluates multiple models including GPT-3.5, and analyzes the challenges and performance in summarizing historical, cross-lingual texts.

Findings

GPT-3.5 achieves moderate to good quality summaries.

Intermediate finetuning models perform poorly.

Longer, older texts are more difficult to summarize.

Abstract

While summarization has been extensively researched in natural language processing (NLP), cross-lingual cross-temporal summarization (CLCTS) is a largely unexplored area that has the potential to improve cross-cultural accessibility and understanding. This paper comprehensively addresses the CLCTS task, including dataset creation, modeling, and evaluation. We (1) build the first CLCTS corpus with 328 instances for hDe-En (extended version with 455 instances) and 289 for hEn-De (extended version with 501 instances), leveraging historical fiction texts and Wikipedia summaries in English and German; (2) examine the effectiveness of popular transformer end-to-end models with different intermediate finetuning tasks; (3) explore the potential of GPT-3.5 as a summarizer; (4) report evaluations from humans, GPT-4, and several recent automatic evaluation metrics. Our results indicate that intermediate task finetuned end-to-end models generate bad to moderate quality summaries while GPT-3.5, as a zero-shot summarizer, provides moderate to good quality outputs. GPT-3.5 also seems very adept at normalizing historical text. To assess data contamination in GPT-3.5, we design an adversarial attack scheme in which we find that GPT-3.5 performs slightly worse for unseen source documents compared to seen documents. Moreover, it sometimes hallucinates when the source sentences are inverted against its prior knowledge with a summarization accuracy of 0.67 for plot omission, 0.71 for entity swap, and 0.53 for plot negation. Overall, our regression results of model performances suggest that longer, older, and more complex source texts (all of which are more characteristic for historical language variants) are harder to summarize for all models, indicating the difficulty of the CLCTS task.