Yonatan Bisk is an Assistant Professor at Carnegie Mellon University (CMU) in the School of Computer Science , with dual appointments in the Language Technologies Institute and Robotics Institute . His research bridges Natural Language Processing (NLP) with robotics, focusing on grounded and embodied language understanding. Assistant Professor, Language Technologies Institute, CMU (2021–Present) Courtesy Appointment, Robotics Institute, CMU Research Themes : Language as a social codification of embodied experience Interpretable multimodal model training Human-robot collaboration frameworks Embodied question-answering systems Selected Trends : His recent publications show increasing focus on cross-modal attention mechanisms (Vid2Robot), error detection in toolchains (Tools Fail), and theory-of-mind reasoning in language agents (SOTOPIA). Multimodal integration spans vision, audio, and robotic control contexts (ANAVI). Labs & Collaborations : Founder of CLAW Lab (Connecting Language to Action and the World) Collaborations with Microsoft Research, Meta Inc, and CMU's REAL (Robotics, Embodied AI, Learning) community
Raquel Fernández is Full Professor of Computational Linguistics and Dialogue Systems at the University of Amsterdam, where she leads the Dialogue Modelling Group at the Institute for Logic, Language & Computation (ILLC). As Vice-Director for Research at ILLC and a Fellow of the ELLIS Society, she bridges computational linguistics, cognitive science, and artificial intelligence through her research on language use in multimodal and conversational contexts. PhD in Computational Linguistics from King's College London Prior research positions at University of Potsdam and Stanford University's CSLI Her work explores how cognitive constraints, social interaction, and perception shape language use, with a focus on: Visually-grounded language processing Multimodal dialogue modeling Model uncertainty and calibration Language grounding in multimodal data Language learning and semantic change Dialogue reference resolution Recent publications analyze multimodal reasoning limitations, cross-lingual knowledge consistency, and uncertainty modeling in dialogue systems. She has received multiple accolades including an ERC Consolidator Grant , NWO VENI/VIDI/Aspasia fellowships , and EMNLP/GenBench awards . Outstanding Paper Award (EMNLP 2023) Best Data Award (GenBench Workshop 2023) ELLIS Society Fellow ERC Consolidator Grant #819455 recipient NWO VENI/VIDI/Aspasia awardee As a leader in academic service, she serves on the SIGDAT Executive Committee and chairs multiple conference committees. Her lab develops models for multimodal dialogue, visual storytelling, and grounded language understanding.
Teruko Mitamura is a prominent researcher at Carnegie Mellon University with over three decades of contributions to natural language processing, computational linguistics, and artificial intelligence. Her work spans from foundational research in event representation to advanced applications in multimodal systems and question answering. Her research interests focus on event detection and understanding, question answering systems, information retrieval, and multimodal processing. She has made significant contributions to event coreference resolution, timeline construction, and cross-document event analysis, developing methodologies that have become standard in the field. Her work often bridges theoretical advances with practical applications, particularly in complex information environments requiring deep semantic understanding. Natural Language Processing : Specializing in event extraction, coreference resolution, and narrative understanding with over 179 publications Question Answering Systems : Developing advanced techniques for complex question answering, particularly through NTCIR QA Lab and PoliInfo tasks Multimodal Processing : Integrating textual, visual, and temporal information for richer understanding in systems like ProMQA Evaluation Methodologies : Creating robust frameworks for assessing NLP systems through TAC KBP Event Tracks Her recent publication trends show a strong focus on leveraging large language models for event understanding, multimodal question answering, and timeline construction. She has expanded her research into specialized domains including patent analysis and novelty examination, demonstrating the breadth of her research impact across academic and practical applications. Active participant in major NLP conferences including ACL, EMNLP, NAACL, and AAAI with consistent publications Long-standing collaborator with researchers at CMU's Language Technologies Institute including Eduard H. Hovy and Eric Nyberg Contributor to shared tasks that have shaped research directions in event processing and question answering Organizer of multiple NTCIR QA Lab tasks focused on political information question answering Dr. Mitamura has mentored numerous researchers who have gone on to make their own contributions to the field, as evidenced by her extensive co-authorship network and the progression of her former students and collaborators into faculty and research positions. Her work continues to evolve with the field while maintaining her focus on deep semantic understanding of events and narratives.
Bryan A. Plummer is an Assistant Professor in the Department of Computer Science at Boston University, affiliated with the IVC Group and the Artificial Intelligence Research (AIR) initiative at the Rafik B. Hariri Institute. He holds a PhD from the University of Illinois at Urbana-Champaign, specializing in computer vision. His research focuses on multimodal machine learning, efficient neural architectures, explainable AI, and robust ML systems. Plummer's work bridges vision and language, addressing challenges in domain generalization, synthetic data utilization, and model efficiency. Notable contributions include the Flickr30K Entities dataset and advancements in vision-language model robustness against web artifacts. He has advised over 20 students, with several securing roles at top institutions like NVIDIA and Google. His recent awards include the 3M Foundation Fellowship and NSF GRFP honorable mention. Plummer actively serves on conference committees (NeurIPS, CVPR, ICCV) and leads initiatives like the 1st Findings Workshop at ICCV'25.
Wenhu Chen is an Assistant Professor at the University of Waterloo's Computer Science Department and a CIFAR AI Chair at the Vector Institute. He also holds a part-time role as a Senior Research Scientist at Google DeepMind (20% allocation). His research focuses on natural language processing, deep learning, and multimodal reasoning, with contributions to models like MAmmoTH, OpenCoderInterpreter, and VISTA. He received awards including the Canada CIFAR AI Chair (2022) and the UCSB CS Outstanding Dissertation Award (2021). Education: PhD in Computer Science from the University of California, Santa Barbara (under William Wang and Xifeng Yan). Research interests include complex reasoning, controllable GenAI, and multimodal benchmarks like MEGABench and MMMU. Grants include CIFAR AI Chair Funding (2022-2027), NSERC Discovery Fund (2023-2028), and multiple NRC Canada grants. He directs the TIGER Lab, advancing generative models in text, images, videos, and music. Recent talks include presentations on multimodal reasoning at Apple and NeurIPS workshops.
Prof. Anya Belz is Full Professor of Computer Science at Dublin City University's School of Computing and Science Lead at ADAPT Research Centre. A leading NLP researcher with PhD-level expertise, she specializes in natural language generation, evaluation methodologies, and multimodal systems. Recipient of multiple best paper awards and NAACL Test of Time Award nomination. Research innovations include foundational work on statistical language generation (deployed in weather forecasting systems), comparative evaluation frameworks, vision-language integration, and reproducibility quantification. Current EPSRC-funded ReproHum project coordinates 20 global labs studying evaluation consistency. Achievements : Developed industry-deployed generation systems for accessibility applications Pioneered cross-modal alignment techniques for image description Authored 100+ publications spanning generation, evaluation, and reproducibility
Bettina Migge is a Full Professor in the School of Languages, Cultures and Linguistics at University College Dublin, where she has been a faculty member since 2004. She previously held academic positions at Goethe University Frankfurt and earned her PhD in Linguistics from The Ohio State University. She served as Head of School from 2016 to 2021 and has held leadership roles in research centers and committees, including the Royal Irish Academy and the Society for Pidgin and Creole Linguistics, of which she is President until 2023. She is Co-Editor of the Journal of Pidgin and Creole Languages and actively contributes to COST Action LITHME, focusing on language and technology. Her educational background includes studies at Universität Hamburg, Université de Cameroun, Freie Universität Berlin, and The Ohio State University, culminating in a PhD focused on African languages in creole genesis, with fieldwork in Suriname and Benin. She is fluent in Dutch, English, French, German, and Ndyuka/Pamaka. Bettina Migge's research centers on sociolinguistics, language contact, and creole languages, with a focus on multilingual contexts undergoing rapid social change such as urbanization and migration. Her work spans French Guiana, Suriname, and Ireland, examining structural and socio-pragmatic aspects of language use, linguistic landscaping, and computer-mediated communication. She has led significant projects like the trilingual DicoNenge(e) dictionary and has explored the role of AI in reshaping language practices. Her recent publications reflect a critical engagement with the impact of AI and machine learning on language, revealing colonial continuities in language technologies and questioning dataist ideologies. She investigates how language is materially produced in both colonial and digital contexts, emphasizing power, authenticity, and control. She has received research funding from the National Science Foundation (USA), IRCHSS, IRC, Ulysses grants, and French research units like SeDyL. Her collaborative projects include work on complementation in creoles, South Dublin English, and language practices in multilingual border zones. Bettina Migge supervises MA and PhD students in sociolinguistics, linguistic landscaping, and World Englishes. She is involved in numerous editorial and professional associations, including the Society for Caribbean Linguistics and The Global Council on Anthropological Linguistics. She has coordinated modules such as World Englishes, Sociolinguistics, and Research in Creole Languages. She leads and participates in public engagement activities, including workshops on AI in linguistics, multilingualism in Ireland, and dictionary launches. Her research team includes collaborators from Europe and beyond, and she is committed to participatory and ethnographically grounded approaches in language documentation and analysis.
Vaishak Belle is a Reader in Logic and Learning at the School of Informatics, University of Edinburgh, and serves as Director of Research and Innovation at the Bayes Centre (40% position). His academic career focuses on the critical intersection of artificial intelligence, formal logic, and machine learning systems. His research spans multiple domains within AI: Neurosymbolic AI integration approaches Logic-based machine learning frameworks Ethics, fairness, and responsibility in AI systems Generative AI and its societal impact Robotics with advanced reasoning capabilities Causal modeling and probabilistic reasoning Belle leads a research laboratory dedicated to advancing neurosymbolic AI, which combines the explainability of symbolic systems with the learning capabilities of neural networks. His recent work has explored abduction in logical frameworks, formal languages for AI safety, and the relevance of logic for general-purpose AI systems. He has made significant contributions to bridging the gap between theoretical AI foundations and practical applications. His publications demonstrate consistent focus on creating AI systems that can reason formally while learning from data, with applications ranging from robotics to ethical decision-making frameworks. His research addresses fundamental challenges in developing trustworthy, explainable AI that can operate safely in complex environments. Beyond technical research, Belle is actively engaged in AI education through executive programs on AI leadership and ethics, and has contributed to public understanding through his children's book "The girl and the robot" which explores human-robot relationships and emotional projection onto artificial agents. He maintains an active presence in the international AI research community, regularly participating in major workshops including Dagstuhl seminars on neurosymbolic AI, and collaborating with researchers across disciplines from computer science to philosophy.
Devis Tuia serves as Associate Professor at the Swiss Federal Institute of Technology Lausanne (EPFL), holding appointments in the Institute of Environmental Engineering (IIE) within the School of Architecture, Civil and Environmental Engineering (ENAC). He leads the Environmental Computational Science and Earth Observation Laboratory (ECEO) since 2020 and contributes to EPFL's Doctoral Program in Civil and Environmental Engineering. His academic journey began in Lausanne with studies at UNIL and EPFL, culminating in a PhD in remote sensing from UNIL. Postdoctoral research followed at institutions in Valencia, Boulder, and EPFL, focusing on machine learning model adaptation. He progressed from Research Assistant Professor at University of Zurich to Associate and Full Professor at Wageningen University before joining EPFL. Tuia's research bridges Earth observation with artificial intelligence, specializing in interpretable deep learning for environmental applications. His lab develops algorithms for making remote sensing accessible, with particular emphasis on digital wildlife conservation through automated censuses using drone and satellite imagery. Current projects tackle the 'black box' problem in environmental modeling while advancing spatial intelligence for sustainable urban development. His 2023-2025 publication portfolio reveals three dominant trends: (1) species distribution modeling using incomplete observations, (2) multimodal fusion of satellite/drone data with textual descriptions, and (3) interpretable AI frameworks for environmental decision-making. This work consistently addresses real-world challenges like wildfire forecasting and biodiversity monitoring. As an educator, Tuia supervises 12 current PhD students and has graduated 4 former EPFL doctoral candidates. His teaching portfolio includes Frontiers of Deep Learning for Engineers , Sensing and Spatial Modeling for Earth Observation , and Image Processing for Earth Observation courses. The ECEO laboratory maintains active collaborations with ESA-NASA initiatives and conservation organizations globally.
Nils Holzenberger is an Assistant Professor at Télécom Paris, France, since February 2023, affiliated with the Data Intelligence Graphs (DIG) research team within the Information Processing and Communication Laboratory (LTcI). His work bridges artificial intelligence, natural language processing, and legal domains through neuro-symbolic approaches to statutory reasoning, particularly in tax law. Education: PhD in Computer Science, Johns Hopkins University (2017-2022) Master's in Engineering, Mines ParisTech (2013-2017) Preparatory Classes, Lycée Louis-le-Grand (2011-2013) Holzenberger's research centers on legal artificial intelligence with emphasis on statutory reasoning limitations in large language models. He pioneered the SARA dataset for tax law reasoning and LegalBench benchmark, developing hybrid symbolic-neural frameworks that expose LLMs' shortcomings in precise legal interpretation. His work integrates Prolog solvers with NLP techniques to create executable tax code mappings and contract analysis tools, establishing foundational methods for verifiable legal AI systems. Analysis of his 15 most recent publications (2019-2024) reveals three dominant research thrusts: (1) Tax law reasoning benchmarks exposing LLM hallucinations, (2) Neuro-symbolic integration for statutory interpretation, and (3) Low-resource template extraction for legal documents. His work consistently demonstrates that pure neural approaches fail at precise legal reasoning, necessitating symbolic grounding for reliable legal AI applications. The DIG research team at Télécom Paris, where Holzenberger leads legal AI initiatives, is actively hiring faculty for neuro-symbolic projects. While specific grant details aren't public, his collaborations with HEC Paris, Copilex startup, and featured podcast appearances indicate substantial industry-academia engagement in legal tech development. Holzenberger directs the legal AI vertical within LTcI's DIG team, focusing on data intelligence for statutory reasoning. His group develops tools for tax minimization strategy discovery, contract analysis, and legal information extraction, maintaining close ties with legal practitioners through projects like the Prolog-based tax code interpreter. The team's infrastructure supports both academic research and startup partnerships in computational law.
Shih-Fu Chang is the Dean of Columbia Engineering and holds the Morris A. and Alma Schapiro Professorship at Columbia University. His research focuses on computer vision, machine learning, and multimedia information retrieval. He is recognized as a foundational figure in the field of content-based visual search and has pioneered innovations in image/video search engines, crime prevention systems, and brain-machine interfaces. His leadership roles include Chair of Columbia's Electrical Engineering Department (2007-2010), Editor-in-Chief of the IEEE Signal Processing Magazine (2006-2008), and Senior Executive Vice Dean at Columbia Engineering, where he drives strategic planning and international collaboration. Dr. Chang has received prestigious awards including the ACM Multimedia Technical Achievement Award, IEEE Signal Processing Technical Achievement Award, and IEEE Kiyo Tomiyasu Award. He is a Fellow of AAAS, ACM, and IEEE, and an Academician of Academia Sinica. His recent work emphasizes multimodal reasoning, few-shot learning, and vision-language systems, with applications in healthcare diagnostics and multimedia benchmarking. His research spans cross-modal understanding, event extraction, and adaptive AI systems. Key contributions include systems like Ferret-v2 for multimodal grounding and RESIN for schema-guided event tracking. He has advised multiple startups and actively contributes to curriculum development in AI and engineering education.
Dylan Hadfield-Menell is an Assistant Professor in the Department of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology, holding the Bonnie and Marty (1964) Tenenbaum Career Development Professorship. His research focuses on AI alignment and human-AI interaction within MIT's School of Engineering. His research interests center on agent alignment problems in AI systems, particularly examining uncertainty in objective optimization for human-robot teams and societal oversight of machine learning systems. Key areas include the principal-agent alignment problem , assistance games frameworks , and robust preference learning that accounts for hidden contextual factors in reinforcement learning from human feedback. His recent publications reveal strong trends toward multi-agent cooperation , formal contract mechanisms for resolving social dilemmas, and advanced evaluation methodologies for AI safety. The research spans theoretical frameworks like open-universe assistance games while addressing practical challenges in language model alignment and cultural bias assessment. Scientific awards include: AI2050 Early Career Fellowship from Schmidt Futures Berkeley Fellowship NSF Graduate Research Fellowship C.V. Ramamoorthy Distinguished Research Award His work bridges theoretical computer science with real-world AI governance challenges, as demonstrated through MIT's participation in AI policy white papers. Current research directions include developing frameworks for transparent AI systems and addressing fundamental limitations in aligning recommender systems with human values through interdisciplinary synthesis.
Jason Eisner is a Professor in the Department of Computer Science at Johns Hopkins University's Whiting School of Engineering, with a secondary joint appointment in Cognitive Science. He is affiliated with the Center for Language and Speech Processing (CLSP), the Human Language Technology Center of Excellence, and leads JHU's cross-departmental machine learning group. His research focuses on developing probabilistic modeling, inference, and learning techniques for linguistic structure. Eisner has authored over 100 papers in computational linguistics, particularly in parsing, grammar induction, machine translation, computational phonology, computational morphology, and weighted finite-state methods. He is the lead designer of Dyna, a declarative programming language for AI research that allows concise programs backed by efficiency tricks. Eisner's work centers on novel methods in NLP and machine learning, with emphasis on probabilistic modeling and inference in complex, structured settings. His research program combines computer science with statistics and linguistics to create statistical models that capture linguistic structure and develop efficient algorithms for applying these models to data with minimal supervision or through large pre-trained models. As an ACL Fellow, Eisner has made significant contributions to the field. His recent work (2023-2025) shows a strong focus on large language models, semantic parsing, controlled text generation, model interpretability, and privacy-preserving techniques, continuing his long-standing interest in the intersection of probabilistic modeling and linguistic structure. ACL Fellow He teaches courses including Natural Language Processing (601.465/665), Machine Learning: Linguistic and Sequence Modeling (601.765), Declarative Methods (601.325/425/625), and Selected Topics in Natural Language Processing (601.865). His advising focuses on research students through the Argo research group, with emphasis on fundamental research questions in NLP rather than immediate applied engineering.
Qingguo Li is a Professor and Associate Head at the Department of Mechanical and Materials Engineering , Queen's University , and a member of the Ingenuity Labs Research Institute . He specializes in biomechanical system design, energy harvesting, wearable sensors, gait analysis, and load carriage systems. His research integrates robotics, biomedical engineering, and sensor technology to develop human-centric devices and mobility aids. Current Roles : Professor, Associate Head, Queen's University Research Institute : Ingenuity Labs Research Institute Lab : Bio-Mechatronics and Robotics Laboratory His work focuses on biomechanical energy harvesting , IMU-based motion analysis , and assistive device development . Key applications include stroke rehabilitation, gait monitoring, and wearable power generation systems. Articles span cable-driven robots , smart walkers , and 3D printing mechanisms , emphasizing human-robot interaction and dynamic modeling . The lab explores sensor calibration , adaptive control algorithms , and human movement optimization . Areas of impact include rehabilitation engineering , load carriage stability , wearable sensor accuracy , and assistive robotics . His team develops solutions for gait asymmetry detection , post-stroke mobility , and low-cost energy systems , leveraging machine learning and kinetic modeling .
Shang-Tse Chen is an Associate Professor at the Department of Computer Science and Information Engineering and Graduate Institute of Networking and Multimedia , National Taiwan University . He leads the NTU AI Security Lab , focusing on applied and theoretical machine learning with emphasis on cybersecurity, adversarial ML, and ML privacy/fairness. Education: PhD in Computer Science (Georgia Tech, 2019), BSc in CSIE (NTU, 2010) Awards: K. T. Li Young Researcher Award (2025), IBM PhD Fellowship (2018), KDD Best Student Paper Runner-Up (2016), NSF SaTC Grant (2017-2021) His research spans adversarial ML, certified defenses, model inversion attacks, and intersection with differential privacy/fairness. Recent work includes physical adversarial attacks on object detectors and practical defenses using JPEG compression. He teaches courses like Security and Privacy of Machine Learning and Introduction to Medical Informatics . Key publication trends show focus on adversarial robustness (ICML/NeurIPS/ICLR), cybersecurity applications (ACSAC), and ML fairness (ACL/EMNLP). Collaborations include industry partnerships with Intel Labs and Symantec. Scientific Awards: K. T. Li Young Researcher Award (2025) ACM TiiS Best Paper Honorable Mention (2020) IBM PhD Fellowship (2018) KDD Audience Appreciation Award Runner-Up (2018) Symantec Fellowship Runner-Up (2016) KDD Best Student Paper Runner-Up (2016) NSF Grant (2017) He advises 13 current students (PhD/MS/Undergrad) and has mentored alumni now at CMU/UC Berkeley. The lab actively recruits postdocs and students across levels.
