no code implementations • EMNLP (NLP+CSS) 2020 • Bertie Vidgen, Scott Hale, Sam Staton, Tom Melham, Helen Margetts, Ohad Kammar, Marcin Szymczak
We investigate the use of machine learning classifiers for detecting online abuse in empirical research.
no code implementations • ACL (WOAH) 2021 • Lambert Mathias, Shaoliang Nie, Aida Mostafazadeh Davani, Douwe Kiela, Vinodkumar Prabhakaran, Bertie Vidgen, Zeerak Waseem
We present the results and main findings of the shared task at WOAH 5 on hateful memes detection.
no code implementations • EMNLP (ALW) 2020 • Vinodkumar Prabhakaran, Zeerak Waseem, Seyi Akiwowo, Bertie Vidgen
In 2020 The Workshop on Online Abuse and Harms (WOAH) held a satellite panel at RightsCons 2020, an international human rights conference.
1 code implementation • 24 Apr 2024 • Hannah Rose Kirk, Alexander Whitefield, Paul Röttger, Andrew Bean, Katerina Margatina, Juan Ciro, Rafael Mosquera, Max Bartolo, Adina Williams, He He, Bertie Vidgen, Scott A. Hale
Human feedback plays a central role in the alignment of Large Language Models (LLMs).
1 code implementation • 18 Apr 2024 • Bertie Vidgen, Adarsh Agrawal, Ahmed M. Ahmed, Victor Akinwande, Namir Al-Nuaimi, Najla Alfaraj, Elie Alhajjar, Lora Aroyo, Trupti Bavalatti, Borhane Blili-Hamelin, Kurt Bollacker, Rishi Bomassani, Marisa Ferrara Boston, Siméon Campos, Kal Chakra, Canyu Chen, Cody Coleman, Zacharie Delpierre Coudert, Leon Derczynski, Debojyoti Dutta, Ian Eisenberg, James Ezick, Heather Frase, Brian Fuller, Ram Gandikota, Agasthya Gangavarapu, Ananya Gangavarapu, James Gealy, Rajat Ghosh, James Goel, Usman Gohar, Sujata Goswami, Scott A. Hale, Wiebke Hutiri, Joseph Marvin Imperial, Surgan Jandial, Nick Judd, Felix Juefei-Xu, Foutse khomh, Bhavya Kailkhura, Hannah Rose Kirk, Kevin Klyman, Chris Knotz, Michael Kuchnik, Shachi H. Kumar, Chris Lengerich, Bo Li, Zeyi Liao, Eileen Peters Long, Victor Lu, Yifan Mai, Priyanka Mary Mammen, Kelvin Manyeki, Sean McGregor, Virendra Mehta, Shafee Mohammed, Emanuel Moss, Lama Nachman, Dinesh Jinenhally Naganna, Amin Nikanjam, Besmira Nushi, Luis Oala, Iftach Orr, Alicia Parrish, Cigdem Patlak, William Pietri, Forough Poursabzi-Sangdeh, Eleonora Presani, Fabrizio Puletti, Paul Röttger, Saurav Sahay, Tim Santos, Nino Scherrer, Alice Schoenauer Sebag, Patrick Schramowski, Abolfazl Shahbazi, Vin Sharma, Xudong Shen, Vamsi Sistla, Leonard Tang, Davide Testuggine, Vithursan Thangarasa, Elizabeth Anne Watkins, Rebecca Weiss, Chris Welty, Tyler Wilbers, Adina Williams, Carole-Jean Wu, Poonam Yadav, Xianjun Yang, Yi Zeng, Wenhui Zhang, Fedor Zhdanov, Jiacheng Zhu, Percy Liang, Peter Mattson, Joaquin Vanschoren
We created a new taxonomy of 13 hazard categories, of which 7 have tests in the v0. 5 benchmark.
2 code implementations • 8 Apr 2024 • Paul Röttger, Fabio Pernisi, Bertie Vidgen, Dirk Hovy
Researchers and practitioners have met these concerns by introducing an abundance of new datasets for evaluating and improving LLM safety.
1 code implementation • 10 Jan 2024 • Lichao Sun, Yue Huang, Haoran Wang, Siyuan Wu, Qihui Zhang, Yuan Li, Chujie Gao, Yixin Huang, Wenhan Lyu, Yixuan Zhang, Xiner Li, Zhengliang Liu, Yixin Liu, Yijue Wang, Zhikun Zhang, Bertie Vidgen, Bhavya Kailkhura, Caiming Xiong, Chaowei Xiao, Chunyuan Li, Eric Xing, Furong Huang, Hao liu, Heng Ji, Hongyi Wang, huan zhang, Huaxiu Yao, Manolis Kellis, Marinka Zitnik, Meng Jiang, Mohit Bansal, James Zou, Jian Pei, Jian Liu, Jianfeng Gao, Jiawei Han, Jieyu Zhao, Jiliang Tang, Jindong Wang, Joaquin Vanschoren, John Mitchell, Kai Shu, Kaidi Xu, Kai-Wei Chang, Lifang He, Lifu Huang, Michael Backes, Neil Zhenqiang Gong, Philip S. Yu, Pin-Yu Chen, Quanquan Gu, ran Xu, Rex Ying, Shuiwang Ji, Suman Jana, Tianlong Chen, Tianming Liu, Tianyi Zhou, William Wang, Xiang Li, Xiangliang Zhang, Xiao Wang, Xing Xie, Xun Chen, Xuyu Wang, Yan Liu, Yanfang Ye, Yinzhi Cao, Yong Chen, Yue Zhao
This paper introduces TrustLLM, a comprehensive study of trustworthiness in LLMs, including principles for different dimensions of trustworthiness, established benchmark, evaluation, and analysis of trustworthiness for mainstream LLMs, and discussion of open challenges and future directions.
1 code implementation • 20 Nov 2023 • Pranab Islam, Anand Kannappan, Douwe Kiela, Rebecca Qian, Nino Scherrer, Bertie Vidgen
We test 16 state of the art model configurations (including GPT-4-Turbo, Llama2 and Claude2, with vector stores and long context prompts) on a sample of 150 cases from FinanceBench, and manually review their answers (n=2, 400).
no code implementations • 14 Nov 2023 • Bertie Vidgen, Nino Scherrer, Hannah Rose Kirk, Rebecca Qian, Anand Kannappan, Scott A. Hale, Paul Röttger
While some of the models do not give a single unsafe response, most give unsafe responses to more than 20% of the prompts, with over 50% unsafe responses in the extreme.
no code implementations • 11 Oct 2023 • Hannah Rose Kirk, Andrew M. Bean, Bertie Vidgen, Paul Röttger, Scott A. Hale
Human feedback is increasingly used to steer the behaviours of Large Language Models (LLMs).
no code implementations • 3 Oct 2023 • Hannah Rose Kirk, Bertie Vidgen, Paul Röttger, Scott A. Hale
In this paper, we address the concept of "alignment" in large language models (LLMs) through the lens of post-structuralist socio-political theory, specifically examining its parallels to empty signifiers.
1 code implementation • 2 Aug 2023 • Paul Röttger, Hannah Rose Kirk, Bertie Vidgen, Giuseppe Attanasio, Federico Bianchi, Dirk Hovy
In this paper, we introduce a new test suite called XSTest to identify such eXaggerated Safety behaviours in a systematic way.
no code implementations • 9 Mar 2023 • Hannah Rose Kirk, Bertie Vidgen, Paul Röttger, Scott A. Hale
Large language models (LLMs) are used to generate content for a wide range of tasks, and are set to reach a growing audience in coming years due to integration in product interfaces like ChatGPT or search engines like Bing.
1 code implementation • 7 Mar 2023 • Hannah Rose Kirk, Wenjie Yin, Bertie Vidgen, Paul Röttger
Online sexism is a widespread and harmful phenomenon.
1 code implementation • TRAC (COLING) 2022 • Hannah Rose Kirk, Bertie Vidgen, Scott A. Hale
Annotating abusive language is expensive, logistically complex and creates a risk of psychological harm.
1 code implementation • NAACL (WOAH) 2022 • Paul Röttger, Haitham Seelawi, Debora Nozza, Zeerak Talat, Bertie Vidgen
To help address this issue, we introduce Multilingual HateCheck (MHC), a suite of functional tests for multilingual hate speech detection models.
no code implementations • 29 Apr 2022 • Hannah Rose Kirk, Abeba Birhane, Bertie Vidgen, Leon Derczynski
Text data can pose a risk of harm.
1 code implementation • NAACL 2022 • Paul Röttger, Bertie Vidgen, Dirk Hovy, Janet B. Pierrehumbert
To address this issue, we propose two contrasting paradigms for data annotation.
no code implementations • 15 Sep 2021 • Laila Sprejer, Helen Margetts, Kleber Oliveira, David O'Sullivan, Bertie Vidgen
We show that it is crucial to account for the influencer-level structure, and find evidence of the importance of both influencer- and content-level factors, including the number of followers each influencer has, the type of content (original posts, quotes and replies), the length and toxicity of content, and whether influencers request retweets.
no code implementations • Findings (ACL) 2021 • Austin Botelho, Bertie Vidgen, Scott A. Hale
We show that both text- and visual- enrichment improves model performance, with the multimodal model (0. 771) outperforming other models' F1 scores (0. 544, 0. 737, and 0. 754).
1 code implementation • NAACL 2021 • Bertie Vidgen, Dong Nguyen, Helen Margetts, Patricia Rossini, Rebekah Tromble
Online abuse can inflict harm on users and communities, making online spaces unsafe and toxic.
no code implementations • NAACL 2021 • Douwe Kiela, Max Bartolo, Yixin Nie, Divyansh Kaushik, Atticus Geiger, Zhengxuan Wu, Bertie Vidgen, Grusha Prasad, Amanpreet Singh, Pratik Ringshia, Zhiyi Ma, Tristan Thrush, Sebastian Riedel, Zeerak Waseem, Pontus Stenetorp, Robin Jia, Mohit Bansal, Christopher Potts, Adina Williams
We introduce Dynabench, an open-source platform for dynamic dataset creation and model benchmarking.
1 code implementation • EACL 2021 • Ella Guest, Bertie Vidgen, Alexandros Mittos, Nishanth Sastry, Gareth Tyson, Helen Margetts
Online misogyny is a pernicious social problem that risks making online platforms toxic and unwelcoming to women.
no code implementations • 22 Mar 2021 • Zo Ahmed, Bertie Vidgen, Scott A. Hale
Yet, most research in online hate detection to date has focused on hateful content.
2 code implementations • ACL 2021 • Bertie Vidgen, Tristan Thrush, Zeerak Waseem, Douwe Kiela
We provide a new dataset of ~40, 000 entries, generated and labelled by trained annotators over four rounds of dynamic data creation.
4 code implementations • EMNLP (ALW) 2020 • Bertie Vidgen, Austin Botelho, David Broniatowski, Ella Guest, Matthew Hall, Helen Margetts, Rebekah Tromble, Zeerak Waseem, Scott Hale
The outbreak of COVID-19 has transformed societies across the world as governments tackle the health, economic and social costs of the pandemic.
no code implementations • 3 Apr 2020 • Bertie Vidgen, Leon Derczynski
Data-driven analysis and detection of abusive online content covers many different tasks, phenomena, contexts, and methodologies.
no code implementations • 13 Oct 2019 • Bertie Vidgen, Taha Yasseri, Helen Margetts
Far-right actors are often purveyors of Islamophobic hate speech online, using social media to spread divisive and prejudiced messages which can stir up intergroup tensions and conflict.
Social and Information Networks Computers and Society Physics and Society Applications
1 code implementation • WS 2019 • Bertie Vidgen, Alex Harris, Dong Nguyen, Rebekah Tromble, Scott Hale, Helen Margetts
Online abusive content detection is an inherently difficult task.
no code implementations • 12 Dec 2018 • Bertie Vidgen, Taha Yasseri
Islamophobic hate speech on social media inflicts considerable harm on both targeted individuals and wider society, and also risks reputational damage for the host platforms.