publications

2025

1. Cross-modal augmentation for few-shot multimodal fake news detection

   Ye Jiang , Taihang Wang , Xiaoman Xu , Yimin Wang , Xingyi Song, and Diana Maynard

   Engineering Applications of Artificial Intelligence, 2025

2024

1. Optimising LLM-Driven Machine Translation with Context-Aware Sliding Windows

   Xinye Yang , Yida Mu , Kalina Bontcheva , and Xingyi Song

   Proceedings of the Ninth Conference on Machine Translation, 2024
2. Enhancing Data Quality through Simple De-duplication: Navigating Responsible Computational Social Science Research

   Yida Mu , Mali Jin , Xingyi Song, and Nikolaos Aletras

   EMNLP, 2024
3. Overview of the CLEF-2024 CheckThat! Lab: Check-Worthiness, Subjectivity, Persuasion, Roles, Authorities, and Adversarial Robustness

   Alberto Barrón-Cedeño , Firoj Alam , Julia Maria Struß , Preslav Nakov , Tanmoy Chakraborty , Tamer Elsayed , Piotr Przybyła , Tommaso Caselli , Giovanni Da San Martino , Fatima Haouari , and  others

   International Conference of the Cross-Language Evaluation Forum for European Languages, 2024
4. Sheffieldveraai at semeval-2024 task 4: Prompting and fine-tuning a large vision-language model for binary classification of persuasion techniques in memes

   Charlie Grimshaw , Kalina Bontcheva , and Xingyi Song

   In Proceedings of the 18th International Workshop on Semantic Evaluation (SemEval-2024) , 2024
5. Confidence Regulation Neurons in Language Models

   Alessandro Stolfo , Ben Peng Wu , Wes Gurnee , Yonatan Belinkov , Xingyi Song, Mrinmaya Sachan , and Neel Nanda

   In NeurIPS , 2024
6. Comparison between parameter-efficient techniques and full fine-tuning: A case study on multilingual news article classification

   Olesya Razuvayevskaya , Ben Wu , Joao A Leite , Freddy Heppell , Ivan Srba , Carolina Scarton , Kalina Bontcheva , and Xingyi Song

   Plos one, 2024
7. Examining the Limitations of Computational Rumor Detection Models Trained on Static Datasets

   Yida Mu , Xingyi Song, Kalina Bontcheva , and Nikolaos Aletras

   LREC-COLING, 2024
8. Navigating Prompt Complexity for Zero-Shot Classification: A Study of Large Language Models in Computational Social Science

   Yida Mu , Ben P Wu , William Thorne , Ambrose Robinson , Nikolaos Aletras , Carolina Scarton , Kalina Bontcheva , and Xingyi Song

   LREC-COLING, 2024
9. Examining Temporalities on Stance Detection Towards COVID-19 Vaccination

   Yida Mu , Mali Jin , Kalina Bontcheva , and Xingyi Song

   LREC-COLING, 2024
10. Large Language Models Offer an Alternative to the Traditional Approach of Topic Modelling

    Yida Mu , Chun Dong , Kalina Bontcheva , and Xingyi Song

    LREC-COLING, 2024
11. Identifying and Aligning Medical Claims Made on Social Media with Medical Evidence

    Anthony Hughes , and Xingyi Song

    LREC-COLING, 2024

2023

1. Classification-Aware Neural Topic Model Combined with Interpretable Analysis-for Conflict Classification

   Tianyu Liang , Yida Mu , Soonho Kim , Darline Kuate , Julie Lang , Rob Vos , and Xingyi Song

   In Proceedings of the 14th International Conference on Recent Advances in Natural Language Processing , 2023
2. Bio-sieve: Exploring instruction tuning large language models for systematic review automation

   Ambrose Robinson , William Thorne , Ben P Wu , Abdullah Pandor , Munira Essat , Mark Stevenson , and Xingyi Song

   arXiv preprint arXiv:2308.06610, 2023
3. Finding Already Debunked Narratives via Multistage Retrieval: Enabling Cross-Lingual, Cross-Dataset and Zero-Shot Learning

   Iknoor Singh , Carolina Scarton , Xingyi Song, and Kalina Bontcheva

   arXiv preprint arXiv:2308.05680, 2023
4. Sheffieldveraai at semeval-2023 task 3: Mono and multilingual approaches for news genre, topic and persuasion technique classification

   Ben Wu , Olesya Razuvayevskaya , Freddy Heppell , João A Leite , Carolina Scarton , Kalina Bontcheva , and Xingyi Song

   In Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023) , 2023
5. Don’t waste a single annotation: improving single-label classifiers through soft labels

   Ben Wu , Yue Li , Yida Mu , Carolina Scarton , Kalina Bontcheva , and Xingyi Song

   In Findings of the Association for Computational Linguistics: EMNLP 2023 , Dec 2023

   Abs

   In this paper, we address the limitations of the common data annotation and training methods for objective single-label classification tasks. Typically, when annotating such tasks annotators are only asked to provide a single label for each sample and annotator disagreement is discarded when a final hard label is decided through majority voting. We challenge this traditional approach, acknowledging that determining the appropriate label can be difficult due to the ambiguity and lack of context in the data samples. Rather than discarding the information from such ambiguous annotations, our soft label method makes use of them for training. Our findings indicate that additional annotator information, such as confidence, secondary label and disagreement, can be used to effectively generate soft labels. Training classifiers with these soft labels then leads to improved performance and calibration on the hard label test set.

6. VaxxHesitancy: A Dataset for Studying Hesitancy Towards COVID-19 Vaccination on Twitter

   Yida Mu , Mali Jin , Charlie Grimshaw , Carolina Scarton , Kalina Bontcheva , and Xingyi Song

   In Proceedings of the International AAAI Conference on Web and Social Media , Dec 2023
7. GATE Teamware 2: An open-source tool for collaborative document classification annotation

   David Wilby , Twin Karmakharm , Ian Roberts , Xingyi Song, and Kalina Bontcheva

   In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics , May 2023

   Abs

   We present GATE Teamware 2: an open-source web-based platform for managing teams of annotators working on document classification tasks. GATE Teamware 2 is an entirely re-engineered successor to GATE Teamware, using contemporary web frameworks. The software allows the management of teams of multiple annotators, project managers and administrators - including the management of annotators - across multiple projects. Projects can be configured to control and monitor the annotation statistics and have a highly flexible JSON-configurable annotation display which can include arbitrary HTML. Optionally, documents can be uploaded with pre-existing annotations and documents are served to annotators in a random order by default to reduce bias. Crucially, annotators can be trained on applying the annotation guidelines correctly and then screened for quality assurance purposes, prior to being cleared for independent annotation. GATE Teamware 2 can be self-deployed, including in container orchestration environments, or provided as private, hosted cloud instances.GATE Teamware 2 is an open-source software and can be downloaded from \urlhttps://github.com/GATENLP/gate-teamware.A demonstration video of the system has also been made available at \urlhttps://youtu.be/KoXkuhc4fmM.

8. Similarity-Aware Multimodal Prompt Learning for fake news detection

   Ye Jiang , Xiaomin Yu , Yimin Wang , Xiaoman Xu , Xingyi Song, and Diana Maynard

   Information Sciences, May 2023

   Abs

   The standard paradigm for fake news detection relies on utilizing text information to model the truthfulness of news. However, the subtle nature of online fake news makes it challenging to solely rely on textual information for debunking. Recent studies that focus on multimodal fake news detection have demonstrated superior performance compared with text-only methods, thereby establishing a new paradigm for detecting fake news. However, this paradigm may require a large number of training instances or updating the entire set of pre-trained model parameters. Furthermore, existing multimodal approaches typically integrate cross-modal features without considering the potential introduction of noise from unrelated semantic representations. To address these issues, this paper proposes the Similarity-Aware Multimodal Prompt Learning (SAMPLE) framework. Incorporating prompt learning into multimodal fake news detection, we used three prompt templates with a soft verbalizer to detect fake news. Moreover, we introduced a similarity-aware fusing method, which adaptively fuses the intensity of multimodal representation so as to mitigate noise injection from uncorrelated cross-modal features. Evaluation results show that SAMPLE outperformed previous work, achieving higher F1 and accuracy scores on two multimodal benchmark datasets, demonstrating its feasibility in real-world scenarios, regardless of data-rich or few-shot settings.

9. A Large-Scale Comparative Study of Accurate COVID-19 Information versus Misinformation

   Yida Mu , Ye Jiang , Freddy Heppell , Iknoor Singh , Carolina Scarton , Kalina Bontcheva , and Xingyi Song

   ICWSM TrueHealth Workshop, May 2023
10. Classifying COVID-19 Vaccine Narratives

    Yue Li , Carolina Scarton , Xingyi Song, and Kalina Bontcheva

    In Proceedings of the 14th International Conference on Recent Advances in Natural Language Processing , May 2023
11. Categorising Fine-to-Coarse Grained Misinformation: An Empirical Study of the COVID-19 Infodemic

    Ye Jiang , Xingyi Song, Carolina Scarton , Iknoor Singh , Ahmet Aker , and Kalina Bontcheva

    In Proceedings of the 14th International Conference on Recent Advances in Natural Language Processing , May 2023

2022

1. An exploratory study on utilising the web of linked data for product data mining

   Ziqi Zhang , and Xingyi Song

   SN Computer Science, May 2022
2. Comparative Analysis of Engagement, Themes, and Causality of Ukraine-Related Debunks and Disinformation

   Iknoor Singh , Kalina Bontcheva , Xingyi Song, and Carolina Scarton

   In International Conference on Social Informatics , May 2022

2021

1. Text mining occupations from the mental health electronic health record: a natural language processing approach using records from the Clinical Record Interactive Search (CRIS) platform in south London, UK

   Natasha Chilman , Xingyi Song, Angus Roberts , Esther Tolani , Robert Stewart , Zoe Chui , Karen Birnie , Lisa Harber-Aschan , Billy Gazard , David Chandran , and  others

   BMJ open, May 2021
2. Classification aware neural topic model for COVID-19 disinformation categorisation

   Xingyi Song, Johann Petrak , Ye Jiang , Iknoor Singh , Diana Maynard , and Kalina Bontcheva

   PloS one, May 2021

2020

1. Using ontologies to map between research data and policymakers’ presumptions: the experience of the KNOWMAK project

   Diana Maynard , Benedetto Lepori , Johann Petrak , Xingyi Song, and Philippe Laredo

   Scientometrics, May 2020
2. Using deep neural networks with intra-and inter-sentence context to classify suicidal behaviour

   Xingyi Song, Johnny Downs , Sumithra Velupillai , Rachel Holden , Maxim Kikoler , Kalina Bontcheva , Rina Dutta , and Angus Roberts

   In Proceedings of the Twelfth Language Resources and Evaluation Conference , May 2020
3. RP-DNN: A Tweet Level Propagation Context Based Deep Neural Networks for Early Rumor Detection in Social Media

   Jie Gao , Sooji Han , Xingyi Song, and Fabio Ciravegna

   In Proceedings of the Twelfth Language Resources and Evaluation Conference , May 2020
4. Comparing topic-aware neural networks for bias detection of news

   Y Jiang , Y Wang , X Song , and D Maynard

   In Proceedings of 24th European Conference on Artificial Intelligence (ECAI 2020) , May 2020

2019

1. A Python script for adaptive layout optimization of trusses

   Linwei He , Matthew Gilbert , and Xingyi Song

   Structural and Multidisciplinary Optimization, May 2019
2. Team bertha von suttner at semeval-2019 task 4: Hyperpartisan news detection using elmo sentence representation convolutional network

   Ye Jiang , Johann Petrak , Xingyi Song, Kalina Bontcheva , and Diana Maynard

   In Proceedings of the 13th International Workshop on Semantic Evaluation , May 2019
3. Online abuse of UK MPs from 2015 to 2019

   Mark A Greenwood , Mehmet E Bakir , Genevieve Gorrell , Xingyi Song, Ian Roberts , and Kalina Bontcheva

   arXiv preprint arXiv:1904.11230, May 2019
4. Online abuse of UK MPs from 2015 to 2019

   Mark A Greenwood , Mehmet E Bakir , Genevieve Gorrell , Xingyi Song, Ian Roberts , and Kalina Bontcheva

   arXiv preprint arXiv:1904.11230, May 2019

2018

1. A Deep Neural Network Sentence Level Classification Method with Context Information

   Xingyi Song, Johann Petrak , and Angus Roberts

   In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing , May 2018

2017

1. Comparing attitudes to climate change in the media using sentiment analysis based on latent dirichlet allocation

   Ye Jiang , Xingyi Song, Jackie Harrison , Shaun Quegan , and Diana Maynard

   In Proceedings of the 2017 EMNLP workshop: Natural language processing meets journalism , May 2017

2016

1. Sheffield Systems for the English-Romanian WMT Translation Task

   Frédéric Blain , Xingyi Song, and Lucia Specia

   In Proceedings of the First Conference on Machine Translation: Volume 2, Shared Task Papers , May 2016

2014

1. Data selection for discriminative training in statistical machine translation

   Xingyi Song, Lucia Specia , and Trevor Cohn

   In Proceedings of the 17th Annual Conference of the European Association for Machine Translation , May 2014

2013

1. BLEU Deconstructed: Designing a Better MT Evaluation Metric.

   Xingyi Song, Trevor Cohn , and Lucia Specia

   Int. J. Comput. Linguistics Appl., May 2013

2011

1. Regression and ranking based optimisation for sentence level MT evaluation

   Xingyi Song, and Trevor Cohn

   In Proceedings of the Sixth Workshop on Statistical Machine Translation , May 2011