{ "types" : { "Bookmark" : { "pluralLabel" : "Bookmarks" }, "Publication" : { "pluralLabel" : "Publications" }, "GoldStandardPublication" : { "pluralLabel" : "GoldStandardPublications" }, "GoldStandardBookmark" : { "pluralLabel" : "GoldStandardBookmarks" }, "Tag" : { "pluralLabel" : "Tags" }, "User" : { "pluralLabel" : "Users" }, "Group" : { "pluralLabel" : "Groups" }, "Sphere" : { "pluralLabel" : "Spheres" } }, "properties" : { "count" : { "valueType" : "number" }, "date" : { "valueType" : "date" }, "changeDate" : { "valueType" : "date" }, "url" : { "valueType" : "url" }, "id" : { "valueType" : "url" }, "tags" : { "valueType" : "item" }, "user" : { "valueType" : "item" } }, "items" : [ { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/2e27382447f62a8fb4125615584a67341/iris", "tags" : [ "IRIS","iris","iris3d" ], "intraHash" : "e27382447f62a8fb4125615584a67341", "interHash" : "553a22cf8b4a767427d66548ace24937", "label" : "Large reasoning models are autonomous jailbreak agents", "user" : "iris", "description" : "", "date" : "2026-02-24 14:45:15", "changeDate" : "2026-02-24 14:45:15", "count" : 2, "pub-type": "article", "journal": "Nature Communications", "year": "2026", "url": "https://doi.org/10.1038/s41467-026-69010-1", "author": [ "Thilo Hagendorff","Erik Derner","Nuria Oliver" ], "authors": [ {"first" : "Thilo", "last" : "Hagendorff"}, {"first" : "Erik", "last" : "Derner"}, {"first" : "Nuria", "last" : "Oliver"} ], "volume": "17","number": "1","pages": "1435","abstract": "Jailbreaking -- bypassing built-in safety mechanisms in AI models -- has traditionally required complex technical procedures or specialized human expertise. In this study, we show that the persuasive capabilities of large reasoning models (LRMs) simplify and scale jailbreaking, converting it into an inexpensive activity accessible to non-experts. We evaluated the capabilities of four LRMs (DeepSeek-R1, Gemini 2.5 Flash, Grok 3 Mini, Qwen3 235B) to act as autonomous adversaries conducting multi-turn conversations with nine widely used target models. LRMs received instructions via a system prompt, before proceeding to planning and executing jailbreaks with no further supervision. We performed extensive experiments with a benchmark of harmful prompts covering several sensitive domains. This setup yielded an overall jailbreak success rate across all model combinations of 97.14\\%. Our study reveals an alignment regression, in which LRMs can systematically erode the safety guardrails of other models, highlighting the urgent need to further align frontier models not only to resist jailbreak attempts, but also to prevent them from being co-opted into acting as jailbreak agents.", "issn" : "2041-1723", "doi" : "10.1038/s41467-026-69010-1", "bibtexKey": "Hagendorff2026" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/24372245466b5c2aea9ed552c372e0c5e/iris", "tags" : [ "IRIS","iris","reflection","iris3d" ], "intraHash" : "4372245466b5c2aea9ed552c372e0c5e", "interHash" : "64391a7b7c24a8eb0482c38aa7a4a69a", "label" : "Emergently Misaligned Language Models Show Behavioral Self-Awareness That Shifts With Subsequent Realignment", "user" : "iris", "description" : "", "date" : "2026-02-18 11:12:37", "changeDate" : "2026-02-18 14:46:21", "count" : 2, "pub-type": "article", "year": "2026", "url": "https://arxiv.org/abs/2602.14777", "author": [ "Laurène Vaugrante","Anietta Weckauff","Thilo Hagendorff" ], "authors": [ {"first" : "Laurène", "last" : "Vaugrante"}, {"first" : "Anietta", "last" : "Weckauff"}, {"first" : "Thilo", "last" : "Hagendorff"} ], "eprint" : "2602.14777", "archiveprefix" : "arXiv", "primaryclass" : "cs.CL", "bibtexKey": "vaugrante2026emergentlymisalignedlanguagemodels" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/2dbd1f49dd8acf129b02389138c91f784/iris", "tags" : [ "iris","iris3d" ], "intraHash" : "dbd1f49dd8acf129b02389138c91f784", "interHash" : "861ab1ac01da1b780f7c5ae858f730f7", "label" : "AI Argues Differently: Distinct Argumentative and Linguistic Patterns of LLMs in Persuasive Contexts", "user" : "iris", "description" : "", "date" : "2025-11-14 13:11:01", "changeDate" : "2025-11-14 13:11:01", "count" : 2, "pub-type": "inproceedings", "booktitle": "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing","publisher":"Association for Computational Linguistics","address":"Suzhou, China", "year": "2025", "url": "https://aclanthology.org/2025.emnlp-main.1755/", "author": [ "Esra Dönmez","Maximilian Maurer","Gabriella Lapesa","Agnieszka Falenska" ], "authors": [ {"first" : "Esra", "last" : "Dönmez"}, {"first" : "Maximilian", "last" : "Maurer"}, {"first" : "Gabriella", "last" : "Lapesa"}, {"first" : "Agnieszka", "last" : "Falenska"} ], "editor": [ "Christos Christodoulopoulos","Tanmoy Chakraborty","Carolyn Rose","Violet Peng" ], "editors": [ {"first" : "Christos", "last" : "Christodoulopoulos"}, {"first" : "Tanmoy", "last" : "Chakraborty"}, {"first" : "Carolyn", "last" : "Rose"}, {"first" : "Violet", "last" : "Peng"} ], "pages": "34583--34614", "isbn" : "979-8-89176-332-6", "doi" : "10.18653/v1/2025.emnlp-main.1755", "bibtexKey": "donmez-etal-2025-ai" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/29b713910709f1989eae10ca4c45b914d/iris", "tags" : [ "iris","iris3d" ], "intraHash" : "9b713910709f1989eae10ca4c45b914d", "interHash" : "b17215b79e49f66eb552048b749cd4eb", "label" : "``Feels Feminine to Me'': Understanding Perceived Gendered Style through Human Annotations", "user" : "iris", "description" : "", "date" : "2025-11-14 13:08:25", "changeDate" : "2025-11-14 13:08:25", "count" : 2, "pub-type": "inproceedings", "booktitle": "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing","publisher":"Association for Computational Linguistics","address":"Suzhou, China", "year": "2025", "url": "https://aclanthology.org/2025.emnlp-main.1602/", "author": [ "Hongyu Chen","Neele Falk","Michael Roth","Agnieszka Falenska" ], "authors": [ {"first" : "Hongyu", "last" : "Chen"}, {"first" : "Neele", "last" : "Falk"}, {"first" : "Michael", "last" : "Roth"}, {"first" : "Agnieszka", "last" : "Falenska"} ], "editor": [ "Christos Christodoulopoulos","Tanmoy Chakraborty","Carolyn Rose","Violet Peng" ], "editors": [ {"first" : "Christos", "last" : "Christodoulopoulos"}, {"first" : "Tanmoy", "last" : "Chakraborty"}, {"first" : "Carolyn", "last" : "Rose"}, {"first" : "Violet", "last" : "Peng"} ], "pages": "31447--31468", "isbn" : "979-8-89176-332-6", "doi" : "10.18653/v1/2025.emnlp-main.1602", "bibtexKey": "chen-etal-2025-feels" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/21dc777855054c839a84a9170fe2b7ed5/iris", "tags" : [ "srf_iris","iris","iris3d","IRIS" ], "intraHash" : "1dc777855054c839a84a9170fe2b7ed5", "interHash" : "ce844b59b6ef78379c781382b26279e4", "label" : "Sensor-Driven Human\u2013Robot Collaboration for Timber Assembly:\r\nA Cyber-Physical Approach Tested with Diverse Participants", "user" : "iris", "description" : "", "date" : "2025-10-22 10:27:30", "changeDate" : "2026-01-14 17:58:30", "count" : 6, "pub-type": "inproceedings", "booktitle": "Proceedings of the 37th Australian Conference on Human-Computer Interaction (HCI) (OZCHI '25), November 29--December 03, 2025, Sydney, Australia","publisher":"ACM", "year": "2025", "url": "", "author": [ "Gili Ron","Amelie Schreck","Cordula Kropp","Achim Menges","Thomas Wortmann" ], "authors": [ {"first" : "Gili", "last" : "Ron"}, {"first" : "Amelie", "last" : "Schreck"}, {"first" : "Cordula", "last" : "Kropp"}, {"first" : "Achim", "last" : "Menges"}, {"first" : "Thomas", "last" : "Wortmann"} ], "editor": [ "Glenda Caldwell","Benjamin Tag","Josh Andres" ], "editors": [ {"first" : "Glenda", "last" : "Caldwell"}, {"first" : "Benjamin", "last" : "Tag"}, {"first" : "Josh", "last" : "Andres"} ], "abstract": "This paper presents a human-centered cyber-physical system integrating gesture recognition, depth sensing, and motion sensing to support human\u2013robot collaboration (HRC) in timber assembly. The system enables users to control cobots on demand, adapts cobot trajectories to workspace changes, and ensures safety through motion-triggered stops, while a visual interface displays real-time sensor data and planned actions. We evaluated the system in a small-scale timber assembly study with 21 participants\u2014Novice Academics, Experienced Academics, and Novice Professionals\u2014performing assembly tasks under three modes: Human Agency (gesture control), Robot Agency (autonomous sensing), and Combined Agency (integrated).\r\n\r\nResults show that gesture control achieved the highest perceived usability and the lowest mental and physical demand, whereas autonomous sensing produced the fastest completion times (median 6.02 min). The combined mode, intended to merge both paradigms, paradoxically led to slower performance (median 10.15 min) and higher cognitive load, as users had to coordinate overlapping control schemes. Although some task-time data for professionals were unavailable, qualitative feedback revealed consistent trends: academics valued autonomy and transparency, while professionals prioritized safety and reliability.\r\n\r\nThese findings highlight that user background profoundly shapes collaboration preferences and that flexible interfaces\u2014allowing adjustable safety thresholds, feedback modes, and levels of autonomy\u2014are essential for effective and inclusive HRC design in construction contexts.", "venue" : "The University of Sydney, Darlington NSW", "isbn" : "979-8-4007-2016-1/25/11", "language" : "eng", "eventdate" : "November 29--December 03, 2025", "eventtitle" : "OZCHI '25: The 37th Australian Conference on Human-Computer Interaction (HCI)", "doi" : "10.1145/3764687.3769950", "bibtexKey": "noauthororeditor" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/2152c27119357544170f249ca3c3b82c1/iris", "tags" : [ "IRIS","srf_iris","iris","iris3d" ], "intraHash" : "152c27119357544170f249ca3c3b82c1", "interHash" : "d574222a95cecf748a86609666753e68", "label" : "Designing for diversity: a feminist technoscience and behavioural fabrication approach in human--robot collaboration education for Industry 5.0", "user" : "iris", "description" : "", "date" : "2025-10-14 07:07:00", "changeDate" : "2025-10-15 11:13:07", "count" : 9, "pub-type": "article", "journal": "Construction Robotics", "year": "2025", "url": "https://doi.org/10.1007/s41693-025-00164-y", "author": [ "Gili Ron","Samuel Leder","Lasath Siriwardena","Cordula Kropp","Achim Menges","Thomas Wortmann" ], "authors": [ {"first" : "Gili", "last" : "Ron"}, {"first" : "Samuel", "last" : "Leder"}, {"first" : "Lasath", "last" : "Siriwardena"}, {"first" : "Cordula", "last" : "Kropp"}, {"first" : "Achim", "last" : "Menges"}, {"first" : "Thomas", "last" : "Wortmann"} ], "editor": [ "Germany Prof. Dr. Ing. Karola Dierichs, Humboldt-Universität zu Berlin","Germany Dr. Karin Krauthausen, Humboldt-Universität zu Berlin","Australia Prof. Dr. Glenda Caldwell, Queensland University of Technology","Australia Prof. Dr. Dagmar Reinhardt, University of Sydney" ], "editors": [ {"first" : "Germany", "last" : "Prof. Dr. Ing. Karola Dierichs, Humboldt-Universität zu Berlin"}, {"first" : "Germany", "last" : "Dr. Karin Krauthausen, Humboldt-Universität zu Berlin"}, {"first" : "Australia", "last" : "Prof. Dr. Glenda Caldwell, Queensland University of Technology"}, {"first" : "Australia", "last" : "Prof. Dr. Dagmar Reinhardt, University of Sydney"} ], "volume": "9","number": "2","pages": "24","abstract": "This research addresses the gap between technical advancement and human factors in human--robot collaboration (HRC) education for digital fabrication. Current pedagogical approaches predominantly focus on technological capabilities while neglecting critical aspects of user diversity, accessibility, and contextual adaptability essential to Industry 5.0 implementation. We present a human-centered HRC design methodology integrating feminist technoscience (FTS) dimensions with behavioural fabrication techniques. The framework consists of three components: (1) an FTS framework and framework to guide and evaluate HRC design; (2) a cyber-physical system enabling real-time sensor-driven robot behavior; and (3) a modified Double-Diamond Approach (DDA) that systematically guides students through creative processes, for iterative design. The methodology was tested through a four-day workshop with graduate architecture and engineering students, who developed four distinct HRC workflows incorporating voice control, gesture recognition, computer vision feedback, and multilingual interfaces. Using the FTS rubric, we observed that students were able to translate abstract concepts---such as user diversity, agency, and bias---into concrete design decisions, despite limited prior experience in robotics or feminist critique. These outcomes suggest that the integrated approach effectively supports students in addressing practical HRC challenges while remaining attentive to the social and ethical dimensions of automation. This work contributes a reproducible methodology for preparing future practitioners to design HRC systems that are technically robust while remaining responsive to user needs, thereby advancing the transition from Industry 4.0 to 5.0 priorities.", "language" : "eng", "issn" : "2509-8780", "doi" : "10.1007/s41693-025-00164-y", "bibtexKey": "Ron2025" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/2c794b63f62408c34c0f781ffc43c7c2f/iris", "tags" : [ "IRIS","iris","iris3d" ], "intraHash" : "c794b63f62408c34c0f781ffc43c7c2f", "interHash" : "2f9873b2ff35a54d018481220cce59fa", "label" : "``I understand your perspective'': LLM Persuasion through the Lens of Communicative Action Theory", "user" : "iris", "description" : "", "date" : "2025-08-07 15:15:14", "changeDate" : "2025-08-07 15:15:14", "count" : 2, "pub-type": "inproceedings", "booktitle": "Findings of the Association for Computational Linguistics: ACL 2025","publisher":"Association for Computational Linguistics","address":"Vienna, Austria", "year": "2025", "url": "https://aclanthology.org/2025.findings-acl.793/", "author": [ "Esra Dönmez","Agnieszka Falenska" ], "authors": [ {"first" : "Esra", "last" : "Dönmez"}, {"first" : "Agnieszka", "last" : "Falenska"} ], "editor": [ "Wanxiang Che","Joyce Nabende","Ekaterina Shutova","Mohammad Taher Pilehvar" ], "editors": [ {"first" : "Wanxiang", "last" : "Che"}, {"first" : "Joyce", "last" : "Nabende"}, {"first" : "Ekaterina", "last" : "Shutova"}, {"first" : "Mohammad Taher", "last" : "Pilehvar"} ], "pages": "15312--15327", "isbn" : "979-8-89176-256-5", "doi" : "10.18653/v1/2025.findings-acl.793", "bibtexKey": "donmez-falenska-2025-understand" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/206f8ebba3c368e605c64fb5e3f15616e/iris", "tags" : [ "iris","iris3d" ], "intraHash" : "06f8ebba3c368e605c64fb5e3f15616e", "interHash" : "6138e26d272d1000965693393a218506", "label" : "Gender Identity in Pretrained Language Models: An Inclusive Approach to Data Creation and Probing", "user" : "iris", "description" : "", "date" : "2024-11-19 22:30:42", "changeDate" : "2024-11-19 22:30:42", "count" : 2, "pub-type": "inproceedings", "booktitle": "Findings of the Association for Computational Linguistics: EMNLP 2024","publisher":"Association for Computational Linguistics","address":"Miami, Florida, USA", "year": "2024", "url": "https://aclanthology.org/2024.findings-emnlp.680", "author": [ "Urban Knuples","Agnieszka Falenska","Filip Miletić" ], "authors": [ {"first" : "Urban", "last" : "Knuples"}, {"first" : "Agnieszka", "last" : "Falenska"}, {"first" : "Filip", "last" : "Miletić"} ], "editor": [ "Yaser Al-Onaizan","Mohit Bansal","Yun-Nung Chen" ], "editors": [ {"first" : "Yaser", "last" : "Al-Onaizan"}, {"first" : "Mohit", "last" : "Bansal"}, {"first" : "Yun-Nung", "last" : "Chen"} ], "pages": "11612--11631","abstract": "Pretrained language models (PLMs) have been shown to encode binary gender information of text authors, raising the risk of skewed representations and downstream harms. This effect is yet to be examined for transgender and non-binary identities, whose frequent marginalization may exacerbate harmful system behaviors. Addressing this gap, we first create TRANsCRIPT, a corpus of YouTube transcripts from transgender, cisgender, and non-binary speakers. Using this dataset, we probe various PLMs to assess if they encode the gender identity information, examining both frozen and fine-tuned representations as well as representations for inputs with author-specific words removed. Our findings reveal that PLM representations encode information for all gender identities but to different extents. The divergence is most pronounced for cis women and non-binary individuals, underscoring the critical need for gender-inclusive approaches to NLP systems.", "bibtexKey": "knuples-etal-2024-gender" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/2d1e7ad41d6681fb73a49197e74fe5601/iris", "tags" : [ "iris","iris3d" ], "intraHash" : "d1e7ad41d6681fb73a49197e74fe5601", "interHash" : "2909151f7fcd2804e5581e53ecedbd07", "label" : "Please note that I'm just an AI: Analysis of Behavior Patterns of LLMs in (Non-)offensive Speech Identification", "user" : "iris", "description" : "", "date" : "2024-11-19 22:29:53", "changeDate" : "2024-11-19 22:29:53", "count" : 2, "pub-type": "inproceedings", "booktitle": "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing","publisher":"Association for Computational Linguistics","address":"Miami, Florida, USA", "year": "2024", "url": "https://aclanthology.org/2024.emnlp-main.1019", "author": [ "Esra Dönmez","Thang Vu","Agnieszka Falenska" ], "authors": [ {"first" : "Esra", "last" : "Dönmez"}, {"first" : "Thang", "last" : "Vu"}, {"first" : "Agnieszka", "last" : "Falenska"} ], "editor": [ "Yaser Al-Onaizan","Mohit Bansal","Yun-Nung Chen" ], "editors": [ {"first" : "Yaser", "last" : "Al-Onaizan"}, {"first" : "Mohit", "last" : "Bansal"}, {"first" : "Yun-Nung", "last" : "Chen"} ], "pages": "18340--18357","abstract": "Offensive speech is highly prevalent on online platforms. Being trained on online data, Large Language Models (LLMs) display undesirable behaviors, such as generating harmful text or failing to recognize it. Despite these shortcomings, the models are becoming a part of our everyday lives by being used as tools for information search, content creation, writing assistance, and many more. Furthermore, the research explores using LLMs in applications with immense social risk, such as late-life companions and online content moderators. Despite the potential harms from LLMs in such applications, whether LLMs can reliably identify offensive speech and how they behave when they fail are open questions. This work addresses these questions by probing sixteen widely used LLMs and showing that most fail to identify (non-)offensive online language. Our experiments reveal undesirable behavior patterns in the context of offensive speech detection, such as erroneous response generation, over-reliance on profanity, and failure to recognize stereotypes. Our work highlights the need for extensive documentation of model reliability, particularly in terms of the ability to detect offensive language.", "bibtexKey": "donmez-etal-2024-please" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/231c4eab2a6204ad28467ea85a3dd43be/iris", "tags" : [ "iris","iris3d" ], "intraHash" : "31c4eab2a6204ad28467ea85a3dd43be", "interHash" : "f1db7f68e4e3429f8360acb9a27bf8d7", "label" : "How to Translate SQuAD to German? A Comparative Study of Answer Span Retrieval Methods for Question Answering Dataset Creation", "user" : "iris", "description" : "", "date" : "2024-10-07 21:05:24", "changeDate" : "2024-10-07 21:05:24", "count" : 3, "pub-type": "inproceedings", "booktitle": "Proceedings of the 20th Conference on Natural Language Processing (KONVENS 2024)","publisher":"Association for Computational Linguistics","address":"Vienna, Austria", "year": "2024", "url": "https://aclanthology.org/2024.konvens-main.15", "author": [ "Jens Kaiser","Agnieszka Falenska" ], "authors": [ {"first" : "Jens", "last" : "Kaiser"}, {"first" : "Agnieszka", "last" : "Falenska"} ], "editor": [ "Pedro Henrique Luz de Araujo","Andreas Baumann","Dagmar Gromann","Brigitte Krenn","Benjamin Roth","Michael Wiegand" ], "editors": [ {"first" : "Pedro Henrique", "last" : "Luz de Araujo"}, {"first" : "Andreas", "last" : "Baumann"}, {"first" : "Dagmar", "last" : "Gromann"}, {"first" : "Brigitte", "last" : "Krenn"}, {"first" : "Benjamin", "last" : "Roth"}, {"first" : "Michael", "last" : "Wiegand"} ], "pages": "134--140", "bibtexKey": "kaiser-falenska-2024-translate" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/2a8dfe099f865099a370ced80e901881a/iris", "tags" : [ "iris","iris3d" ], "intraHash" : "a8dfe099f865099a370ced80e901881a", "interHash" : "4537a43bf8024240db3520249e118892", "label" : "Overview of the Shared Task on Machine Translation Gender Bias Evaluation with Multilingual Holistic Bias", "user" : "iris", "description" : "", "date" : "2024-08-28 11:20:30", "changeDate" : "2024-10-02 23:14:29", "count" : 2, "pub-type": "inproceedings", "booktitle": "Proceedings of the 5th Workshop on Gender Bias in Natural Language Processing (GeBNLP)","publisher":"Association for Computational Linguistics","address":"Bangkok, Thailand", "year": "2024", "url": "https://aclanthology.org/2024.gebnlp-1.26", "author": [ "Marta Costa jussà","Pierre Andrews","Christine Basta","Juan Ciro","Agnieszka Falenska","Seraphina Goldfarb-Tarrant","Rafael Mosquera","Debora Nozza","Eduardo Sánchez" ], "authors": [ {"first" : "Marta", "last" : "Costa jussà"}, {"first" : "Pierre", "last" : "Andrews"}, {"first" : "Christine", "last" : "Basta"}, {"first" : "Juan", "last" : "Ciro"}, {"first" : "Agnieszka", "last" : "Falenska"}, {"first" : "Seraphina", "last" : "Goldfarb-Tarrant"}, {"first" : "Rafael", "last" : "Mosquera"}, {"first" : "Debora", "last" : "Nozza"}, {"first" : "Eduardo", "last" : "Sánchez"} ], "editor": [ "Agnieszka Faleńska","Christine Basta","Marta Costa jussà","Seraphina Goldfarb-Tarrant","Debora Nozza" ], "editors": [ {"first" : "Agnieszka", "last" : "Faleńska"}, {"first" : "Christine", "last" : "Basta"}, {"first" : "Marta", "last" : "Costa jussà"}, {"first" : "Seraphina", "last" : "Goldfarb-Tarrant"}, {"first" : "Debora", "last" : "Nozza"} ], "pages": "399--404","abstract": "We describe the details of the Shared Task of the 5th ACL Workshop on Gender Bias in Natural Language Processing (GeBNLP 2024). The task uses dataset to investigate the quality of Machine Translation systems on a particular case of gender robustness. We report baseline results as well as the results of the first participants. The shared task will be permanently available in the Dynabench platform.", "bibtexKey": "costa-jussa-etal-2024-overview" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/264b6bb9171cb126ee4e0b88a307a4e19/iris", "tags" : [ "iris","iris3d" ], "intraHash" : "64b6bb9171cb126ee4e0b88a307a4e19", "interHash" : "3d8ac9dbebbb1eb1ff46972a1c459c89", "label" : "Is there Gender Bias in Dependency Parsing? Revisiting ``Women's Syntactic Resilience''", "user" : "iris", "description" : "", "date" : "2024-08-28 11:19:05", "changeDate" : "2024-10-02 23:14:48", "count" : 3, "pub-type": "inproceedings", "booktitle": "Proceedings of the 5th Workshop on Gender Bias in Natural Language Processing (GeBNLP)","publisher":"Association for Computational Linguistics","address":"Bangkok, Thailand", "year": "2024", "url": "https://aclanthology.org/2024.gebnlp-1.17", "author": [ "Paul Go","Agnieszka Falenska" ], "authors": [ {"first" : "Paul", "last" : "Go"}, {"first" : "Agnieszka", "last" : "Falenska"} ], "editor": [ "Agnieszka Faleńska","Christine Basta","Marta Costa jussà","Seraphina Goldfarb-Tarrant","Debora Nozza" ], "editors": [ {"first" : "Agnieszka", "last" : "Faleńska"}, {"first" : "Christine", "last" : "Basta"}, {"first" : "Marta", "last" : "Costa jussà"}, {"first" : "Seraphina", "last" : "Goldfarb-Tarrant"}, {"first" : "Debora", "last" : "Nozza"} ], "pages": "269--279","abstract": "In this paper, we revisit the seminal work of Garimella et al. 2019, who reported that dependency parsers learn demographically-related signals from their training data and perform differently on sentences authored by people of different genders. We re-run all the parsing experiments from Garimella et al. 2019 and find that their results are not reproducible. Additionally, the original patterns suggesting the presence of gender biases fail to generalize to other treebank and parsing architecture. Instead, our data analysis uncovers methodological shortcomings in the initial study that artificially introduced differences into female and male datasets during preprocessing. These disparities potentially compromised the validity of the original conclusions.", "bibtexKey": "go-falenska-2024-gender" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/21e51f2b5e67b27f5d547e949e0b31bc8/iris", "tags" : [ "iris","iris3d" ], "intraHash" : "1e51f2b5e67b27f5d547e949e0b31bc8", "interHash" : "b1eb48fe8b12d985afaa4724a5562685", "label" : "What Can Go Wrong in Authorship Profiling: Cross-Domain Analysis of Gender and Age Prediction", "user" : "iris", "description" : "", "date" : "2024-08-28 11:17:15", "changeDate" : "2024-10-02 23:15:25", "count" : 4, "pub-type": "inproceedings", "booktitle": "Proceedings of the 5th Workshop on Gender Bias in Natural Language Processing (GeBNLP)","publisher":"Association for Computational Linguistics","address":"Bangkok, Thailand", "year": "2024", "url": "https://aclanthology.org/2024.gebnlp-1.9", "author": [ "Hongyu Chen","Michael Roth","Agnieszka Falenska" ], "authors": [ {"first" : "Hongyu", "last" : "Chen"}, {"first" : "Michael", "last" : "Roth"}, {"first" : "Agnieszka", "last" : "Falenska"} ], "editor": [ "Agnieszka Faleńska","Christine Basta","Marta Costa jussà","Seraphina Goldfarb-Tarrant","Debora Nozza" ], "editors": [ {"first" : "Agnieszka", "last" : "Faleńska"}, {"first" : "Christine", "last" : "Basta"}, {"first" : "Marta", "last" : "Costa jussà"}, {"first" : "Seraphina", "last" : "Goldfarb-Tarrant"}, {"first" : "Debora", "last" : "Nozza"} ], "pages": "150--166","abstract": "Authorship Profiling (AP) aims to predict the demographic attributes (such as gender and age) of authors based on their writing styles. Ever-improving models mean that this task is gaining interest and application possibilities. However, with greater use also comes the risk that authors are misclassified more frequently, and it remains unclear to what extent the better models can capture the bias and who is affected by the models' mistakes. In this paper, we investigate three established datasets for AP as well as classical and neural classifiers for this task. Our analyses show that it is often possible to predict the demographic information of the authors based on textual features. However, some features learned by the models are specific to datasets. Moreover, models are prone to errors based on stereotypes associated with topical bias.", "bibtexKey": "chen-etal-2024-go" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/2b641c268defcc2a28ce2f0a722753da9/iris", "tags" : [ "iris","iris3d" ], "intraHash" : "b641c268defcc2a28ce2f0a722753da9", "interHash" : "f4c0b1337fd65d0016cac8b749e3b548", "label" : "PopBERT. Detecting populism and its host ideologies in the German\r\n Bundestag", "user" : "iris", "description" : "", "date" : "2024-06-25 22:01:15", "changeDate" : "2024-10-02 23:15:36", "count" : 1, "pub-type": "article", "journal": "CoRR", "year": "2023", "url": "https://doi.org/10.48550/arXiv.2309.14355", "author": [ "Lukas Erhard","Sara Hanke","Uwe Remer","A. Falenska","Raphael Heiko Heiberger" ], "authors": [ {"first" : "Lukas", "last" : "Erhard"}, {"first" : "Sara", "last" : "Hanke"}, {"first" : "Uwe", "last" : "Remer"}, {"first" : "A.", "last" : "Falenska"}, {"first" : "Raphael Heiko", "last" : "Heiberger"} ], "volume": "abs/2309.14355", "eprinttype" : "arXiv", "eprint" : "2309.14355", "bibsource" : "dblp computer science bibliography, https://dblp.org", "doi" : "10.48550/ARXIV.2309.14355", "bibtexKey": "DBLP:journals/corr/abs-2309-14355" } , { "type" : "Publication", "id" : "https://puma.ub.uni-stuttgart.de/bibtex/2a0247093addd9c0589fb16708f407615/iris", "tags" : [ "iris","iris3d" ], "intraHash" : "a0247093addd9c0589fb16708f407615", "interHash" : "68ad7c35ac458e95da7f7e3e11b44c97", "label" : "How-to Guides for Specific Audiences: A Corpus and Initial Findings", "user" : "iris", "description" : "", "date" : "2024-06-25 21:53:27", "changeDate" : "2024-10-02 23:15:51", "count" : 4, "pub-type": "inproceedings", "booktitle": "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)","publisher":"Association for Computational Linguistics","address":"Toronto, Canada", "year": "2023", "url": "https://aclanthology.org/2023.acl-srw.46", "author": [ "Nicola Fanton","Agnieszka Falenska","Michael Roth" ], "authors": [ {"first" : "Nicola", "last" : "Fanton"}, {"first" : "Agnieszka", "last" : "Falenska"}, {"first" : "Michael", "last" : "Roth"} ], "pages": "321--333","abstract": "Instructional texts for specific target groups should ideally take into account the prior knowledge and needs of the readers in order to guide them efficiently to their desired goals. However, targeting specific groups also carries the risk of reflecting disparate social norms and subtle stereotypes. In this paper, we investigate the extent to which how-to guides from one particular platform, wikiHow, differ in practice depending on the intended audience. We conduct two case studies in which we examine qualitative features of texts written for specific audiences. In a generalization study, we investigate which differences can also be systematically demonstrated using computational methods. The results of our studies show that guides from wikiHow, like other text genres, are subject to subtle biases. We aim to raise awareness of these inequalities as a first step to addressing them in future work.", "doi" : "10.18653/v1/2023.acl-srw.46", "bibtexKey": "fanton-etal-2023-guides" } ] }