Papers by John Pavlopoulos

21 papers
Dialect Normalization using Large Language Models and Morphological Rules (2025.findings-acl)

Copied to clipboard

Challenge: Natural language understanding systems struggle with low-resource languages, including many dialects of high-resourced ones.
Approach: They propose a method that combines rule-based linguistically informed transformations and large language models with targeted few-shot prompting without any parallel data.
Outcome: The proposed method is able to transform dialectal text into a standard variety while maintaining as much of the original meaning as possible.
Learning to Align: Addressing Character Frequency Distribution Shifts in Handwritten Text Recognition (2025.findings-emnlp)

Copied to clipboard

Challenge: Character sets change over time and character frequency distributions shift across historical periods or regions . character distribution alignment can improve existing models at inference time without requiring retraining .
Approach: They propose a loss function that incorporates the Wasserstein distance between predicted and target distributions.
Outcome: The proposed method improves accuracy and robustness under temporal and contextual shifts.
Detecting Erroneously Recognized Handwritten Byzantine Text (2023.findings-emnlp)

Copied to clipboard

Challenge: Handwritten text recognition (HTR) produces textual output that contains errors, which are much higher than recognised printed text.
Approach: They investigate the properties of handwritten texts that lead post-correction systems to this adversarial behaviour in Byzantine Greek.
Outcome: The proposed model achieves an average precision score of 95% in Byzantine Greek and 97% in modern and ancient Greek.
Dating Greek Papyri with Text Regression (2023.acl-long)

Copied to clipboard

Challenge: a large number of Greek papyri documents can only be dated tentatively or in approximation due to the lack of decisive evidence.
Approach: a new study trains regression models to estimate Greek papyri's date using a dataset of 389 transcriptions . the authors propose a method to estimate the date of 159 Greek pamphlets, which are only the upper limit known .
Outcome: a new study predicts a date for Greek papyri with an average MAE of 54 years and an MSE of 1.17 . the model outperforms image classifiers and other baselines for 159 manuscripts, with only the upper limit known .
Deciphering Emotional Landscapes in the Iliad: A Novel French-Annotated Dataset for Emotion Recognition (2024.lrec-main)

Copied to clipboard

Challenge: Using an emotion-annotated dataset, we aim to provide a resource for the scientific community to study the emotional intricacies of classical literature.
Approach: They propose to provide an emotion-annotated dataset for classical literature and Western mythology using a multivariate time series and a deep learning masked language model.
Outcome: The proposed dataset reveals compelling patterns and phenomena within the Iliad's emotional landscape.
Polarized Opinion Detection Improves the Detection of Toxic Language (2024.eacl-long)

Copied to clipboard

Challenge: Existing methods for estimating polarized annotations are un-normalized and difficult to exploit in machine learning.
Approach: They propose a method for K-class text classification that exploits polarized texts in the dataset.
Outcome: The proposed method exploits polarized texts in a dataset and can improve classification performance.
Sentiment Analysis of Homeric Text: The 1st Book of Iliad (2022.lrec-1)

Copied to clipboard

Challenge: Sentiment analysis studies focus more on online customer reviews and social media texts, but are less on literary studies.
Approach: They propose to model the perceived sentiment of Iliad verses using a deep learning masked language model and a pre-trained model to estimate the sentiment of the poem.
Outcome: The proposed model shows that sentiment estimators can be used as mechanical annotators, thus facilitating the distant reading of Homeric text.
Handwritten Paleographic Greek Text Recognition: A Century-Based Approach (2022.lrec-1)

Copied to clipboard

Challenge: achieving high accuracy HTR results for Greek manuscripts is still a major challenge . Optical character recognition software is notoriously difficult to use for handwritten text .
Approach: They propose to use Greek manuscripts as a source for a new model to assess HTR accuracy.
Outcome: The proposed model can be used to improve the recognition rate of Greek manuscripts.
Civil Rephrases Of Toxic Texts With Self-Supervised Transformers (2021.eacl-main)

Copied to clipboard

Challenge: et al., 2018a): a poor phrasing may make the conversation go awry.
Approach: They propose a model that can help suggest rephrasings of toxic comments in a more civil manner.
Outcome: The proposed model generates sentences that are more fluent and better at preserving the initial content compared to earlier systems and human evaluation.
A Data-Driven Guided Decoding Mechanism for Diagnostic Captioning (2024.findings-acl)

Copied to clipboard

Challenge: Diagnostic Captioning (DC) systems receive one or more medical images of a patient, such as X-Rays or Magnetic Resonance Images (MRIs).
Approach: They propose a data-driven guided decoding method that incorporates medical information into the beam search of the diagnostic text generation process.
Outcome: The proposed method improves on two medical datasets and can be used in few- and zero-shot learning scenarios.
Enriching Grammatical Error Correction Resources for Modern Greek (2022.lrec-1)

Copied to clipboard

Challenge: Davidson and Kilgarriff, 2011) have focused on the English language, but there are limited efforts to expand GEC in other languages.
Approach: They develop and test a multilingual text-to-text transformer for Greek . they provide a model that can be fully-fledged for Greek with annotation corrections .
Outcome: The proposed model achieves 52.63% F0.5 on part of the Greek Native Corpus, 16% below the winning system on English GEC.
FoodSafeSum: Enabling Natural Language Processing Applications for Food Safety Document Summarization and Analysis (2025.findings-emnlp)

Copied to clipboard

Challenge: a lack of structured datasets hinders natural language processing research . a new dataset of food safety documents and related metadata is presented .
Approach: They present a dataset of human-written and Large Language Model (LLM)-generated food safety documents . they evaluate their utility on three NLP tasks directly reflecting food safety practices .
Outcome: The proposed dataset performs comparably or better than human summaries on three NLP tasks . it also shows clustering of summary for event tracking and compliance monitoring .
Toxicity Detection: Does Context Really Matter? (2020.acl-main)

Copied to clipboard

Challenge: Existing ‘toxicity’ detection datasets and models ignore the context of the posts, implicitly assuming that comments may be judged independently.
Approach: They limit the notion of context to the previous post in the thread and the discussion title and focus on how it affects human judgement.
Outcome: The proposed model can amplify or mitigate perceived toxicity of posts and a small but significant subset of manually labeled posts end up having the opposite toxicity labels if the annotators are not provided with context.
A Study of Distant Viewing of ukiyo-e prints (2022.lrec-1)

Copied to clipboard

Challenge: ukiyo-e landscape prints feature diverse place-names, both man-made and natural formations.
Approach: They propose to use a Japanese BERT-based Name Entity Recogniser to analyze a visual dataset that is hosted by the Art Research Center at the Ritsumeikan University, Kyoto.
Outcome: The proposed approach improves the work by fine-tuning and applying a Japanese BERT-based Name Entity Recogniser to a visual dataset hosted by the Art Research Center at the Ritsumeikan University, Kyoto.
Evaluation and Facilitation of Online Discussions in the LLM Era: A Survey (2025.emnlp-main)

Copied to clipboard

Challenge: Recent advances in LLMs enable artificial facilitation agents to not only moderate content, but also actively improve the quality of interactions.
Approach: They propose a taxonomy on discussion quality evaluation and a new taxonomies for intervention and facilitation strategies.
Outcome: The proposed methods synthesize ideas from Natural Language Processing (NLP) and Social Sciences to provide a taxonomy on discussion quality evaluation, and a roadmap of good practices and future research directions.
HoLM: Analyzing the Linguistic Unexpectedness in Homeric Poetry (2024.lrec-main)

Copied to clipboard

Challenge: Existing work on the authorship of the Homeric poems has only been done at the level of lengthier excerpts, but not individual verses, at which most suspected interpolations occur.
Approach: They present a corpus of Homeric verses with a score quantifying linguistic unexpectedness based on Perplexity.
Outcome: The proposed corpus of Homeric verses is complemented with a score quantifying linguistic unexpectedness based on Perplexity.
GR-NLP-TOOLKIT: An Open-Source NLP Toolkit for Modern Greek (2025.coling-demos)

Copied to clipboard

Challenge: GR-NLP-TOOLKIT is an open-source natural language processing toolkit for modern Greek.
Approach: They present GR-NLP-TOOLKIT, an open-source natural language processing toolkit for Greek.
Outcome: The toolkit provides state-of-the-art performance in five core NLP tasks . it can be easily installed in Python and is accessible through a demonstration platform on HuggingFace .
From the Detection of Toxic Spans in Online Discussions to the Analysis of Toxic-to-Civil Transfer (2022.acl-long)

Copied to clipboard

Challenge: a dataset of English posts with annotations of toxic spans is released . sequence labeling models perform best, but rationale extraction methods are promising .
Approach: They propose a dataset for toxic spans detection that includes an annotation of toxic posts . they propose to add generic rationale extraction mechanisms to the model to obtain toxic span information .
Outcome: The proposed framework is based on a dataset of English posts with toxic span annotations . it shows that sequence labeling models perform best, but that rationale extraction methods are promising .
CICLe: Conformal In-Context Learning for Largescale Multi-Class Food Risk Classification (2024.findings-acl)

Copied to clipboard

Challenge: Contaminated or adulterated food poses a substantial risk to human health.
Approach: They present a dataset of 7,546 text messages describing public food recalls.
Outcome: The proposed model outperforms RoBERTa and XLM-R on classes with low support while reducing energy consumption.
Still All Greeklish to Me: Greeklish to Greek Transliteration (2024.lrec-main)

Copied to clipboard

Challenge: Greeklish is a writing form that is used to avoid switching languages on multilingual keyboards . even native Greek speakers may struggle to understand Greeklished .
Approach: They propose to use Greeklish to avoid switching languages on multilingual keyboards . they propose to train models on Greek datasets using the Greek alphabet .
Outcome: The proposed model outperforms existing models on Greeklish data.
Towards a Greek Proverb Atlas: Computational Spatial Exploration and Attribution of Greek Proverbs (2024.emnlp-main)

Copied to clipboard

Challenge: a recent study focuses on Greek proverbs, which carry wisdom and are still used today . it is the first large-scale machine-actionable dataset of Greek prowords quantifying their spatial distribution across different locations.
Approach: They propose to use a publicly-available and machine-actionable dataset of Greek proverbs to quantify their spatial distribution across different locations.
Outcome: The proposed dataset is a publicly-available and machine-actionable dataset of Greek proverb variants.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations