Find Academic Supervisors

Qianwen WangView profile

Assistant Professor

• Interactive Visualization
• Interpretable Machine Learning
• Human-AI Collaboration
• +5 more

Qianwen Wang is a tenure-track Assistant Professor in the Computer Science department at the University of Minnesota, Twin Cities. Her research combines interactive visualization with interpretable machine learning to foster intuitive, efficient, and reliable Human-AI collaboration. She actively seeks motivated students, research assistants, and interns to join her dynamic research team at UMN CS. Dr. Wang's research focuses on three primary themes: Human-AI Collaboration, where she designs tools to facilitate Human-AI interaction; Automatic & Intelligent Visualization, where she develops techniques to make visualizations accurately interpreted and easily used; and VIS+(X)AI in Biomed/Healthcare, where she studies how visualization and explainable AI can promote scientific discoveries in biomedicine and healthcare. Her work has made significant contributions to visualization, human-computer interaction, and bioinformatics, with particular applications in biomedical knowledge graphs and single-cell omics analysis. Her recent publications demonstrate a strong trend toward integrating visualization with large language models and graph neural networks for biomedical applications. She has published extensively at top venues including IEEE VIS, ACM CHI, and Nature Medicine, with a focus on making AI systems more interpretable and useful for domain experts, particularly in healthcare contexts. Her work often bridges theoretical advances in visualization with practical applications in genomics and healthcare. Two IEEE VIS Honorable Mention Awards (2022, 2024) Best Paper Award from IMLH@ICML 2021 Two Best Abstract Awards from BioVis ISMB (2021, 2022) HDSI Postdoctoral Research Fund Award Research covered by MIT News and Nature Technology Features Dr. Wang actively contributes to the academic community through service roles including General Chair for ISMB BioVis, VisNotes and Poster Chair for IEEE PacificVis, and Program Committee member for IEEE VIS, ACM CHI, and ACM IUI. Her research has been supported by various grants that enable her team to develop innovative visualization techniques and explore their practical applications in biomedical domains. Her research group maintains an active presence in the visualization and HCI communities, with members participating in major conferences and workshops. The lab focuses on creating tools that bridge the gap between complex AI models and human understanding, with particular emphasis on making these technologies accessible and useful for domain experts in biomedical research.

Sara StymneView profile

Senior Lecturer

• Computational Linguistics
• Natural Language Processing
• Cross-lingual NLP
• +7 more

Sara Stymne is a Senior Lecturer in Computational Linguistics at the Department of Linguistics and Philology, Uppsala University, where she has been working since 2012. She initially joined as a post-doc (2012-2015), then worked as a researcher (2015-2017), and served as an assistant professor (2017-2023) before her current position as Senior Lecturer. Prior to Uppsala, she was a researcher at Linköping University's Department of Computer and Information Science. Dr. Stymne earned her PhD in Computational Linguistics from Linköping University in 2012 with the thesis 'Text Harmonization Strategies for Phrase-Based Statistical Machine Translation,' following a Licentiate degree in Computational Linguistics (2009) and a Master's degree in Cognitive Science (2006), both also from Linköping University. During her doctoral studies, she spent the autumn of 2010 and spring of 2009 at Xerox Research Centre Europe in Grenoble, France. Her primary research interests focus on cross-lingual natural language processing and digital humanities, with particular emphasis on multilingual dependency parsing. Dr. Stymne is passionate about applying computational linguistics to solve research questions in other fields, including language history, literary analysis, and political science. Her earlier work concentrated on machine translation, with specific interests in discourse-aware translation, compound processing, and error analysis. She has made significant contributions to the development of language technology tools for analyzing dialogue, narrative, and stylistic features in literature. Analysis of Dr. Stymne's recent publications reveals a strong focus on cross-lingual and cross-domain natural language processing. Her work spans multiple subfields including dependency parsing across genres and topics, discourse relation analysis in low-resource languages like Egyptian Arabic, direct speech identification in Swedish literature, and causality detection in governmental documents. A notable trend is her application of NLP techniques to digital humanities problems, particularly in analyzing literary texts and historical language change. Her research often involves creating and utilizing specialized datasets for specific linguistic phenomena across multiple languages. Dr. Stymne actively supervises graduate students, having guided numerous master's and bachelor's theses on topics ranging from speech recognition to multilingual parsing and causality detection. She leads or participates in several research projects including 'Fictional prose and language change' (funded by VR, 2021-2023) and 'Enabling climate-resilient development' (funded by Marianne and Marcus Wallenberg Foundation, 2023-2027), demonstrating her commitment to interdisciplinary research with practical applications. Her work has resulted in several notable software resources including uuPronPred for cross-lingual pronoun prediction, uuparser for dependency parsing, and Docent for document-level machine translation. Within the Computational Linguistics and Language Technology group at Uppsala University, Dr. Stymne contributes to multiple research initiatives focused on developing language technology tools for digital humanities applications. Her team works closely with literary scholars and historians to create computational methods for analyzing large corpora of literary texts, particularly focusing on Swedish literature across different historical periods. Her research bridges the gap between theoretical computational linguistics and practical applications in the humanities, creating new methodologies for quantitative analysis of literary and historical texts.

Michael O'BoyleView profile

Professor

• Heterogeneous Code discovery and Optimisation
• Neural Machine Translation and Neural Synthesis
• Deep Neural Network system stack
• +7 more

Michael O'Boyle is a Professor at the University of Edinburgh's School of Informatics, where he serves as Director of the ARM Research Centre of Excellence and the EPSRC Centre for Doctoral Training in Pervasive Parallelism. Holding an EPSRC Established Career Research Fellowship, he leads pioneering work in compiler technology for heterogeneous architectures, bridging theoretical advances with practical high-performance computing applications. Professor O'Boyle's research spans multiple cutting-edge areas including heterogeneous code discovery and optimization, neural machine translation for program synthesis, deep neural network system stack optimization, software-defined hardware, and compiler/architecture co-design. His approach integrates constraint analysis, program synthesis, and machine learning to address complex challenges in high-performance computing across diverse hardware platforms. His recent publications reveal a strong trend toward integrating machine learning with traditional compiler techniques, particularly in neural program synthesis, tensor optimization, and architecture-aware compilation. This work represents a paradigm shift in compiler design, moving from rule-based systems to learning-based approaches that can automatically adapt to diverse hardware targets. IEEE/ACM CGO 2025 Distinguished Paper Award for 'Tensorize: Fast Synthesis of Tensor Programs from Legacy Code' IEEE/ACM CGO 2024 Test of Time Award ACM GPCE 2023 Best Paper Award for 'C2TACO: Lifting Tensor Code to TACOM' ACM ASPLOS 2021 Distinguished Paper Award IEEE HPCA 2021 Best Paper Award for 'Prodigy: Improving the Memory Latency of Data-Indirect Irregular Workloads' Professor O'Boyle has successfully mentored numerous PhD students who have secured prominent positions in academia (including at Cambridge, Edinburgh, Leeds, and McGill) and industry (including Meta, NVIDIA, Qualcomm, Huawei, and Microsoft). His research is supported by significant funding from EPSRC, ARM, and European projects including Bonseyes and Transmuter, demonstrating strong international recognition and industry impact. He leads the influential Compiler and Architecture Design (CArD) Group at the University of Edinburgh and is a founder of the HiPEAC Network of Excellence, which has grown into a major European initiative connecting researchers and practitioners in high-performance and embedded computing.

• Machine Learning
• Deep Learning
• Theoretical Foundations of AI
• +11 more

Mikhail (Misha) Belkin is a Professor at the Halicioglu Data Science Institute (HDSI) at the University of California San Diego , with an affiliated appointment in the Department of Computer Science and Engineering . He is also an Amazon Scholar , reflecting his impactful industry collaboration. Since January 2024, he has served as the Editor-in-Chief of the SIAM Journal on Mathematics of Data Science (SIMODS) . Research Interests: Belkin's research centers on the theoretical foundations of machine learning, particularly the mathematical understanding of modern deep learning. His work investigates interpolation , over-parameterization , and feature learning in neural networks. He is renowned for introducing the double descent risk curve, which reconciles classical bias-variance trade-offs with the success of overfitted models. His recent work identifies the Average Gradient Outer Product (AGOP) as a fundamental mechanism of feature learning, applicable across architectures like CNNs and transformers. Scientific Contributions and Trends: His recent publications, appearing in Science , PNAS , and NeurIPS , demonstrate a strong trend toward unifying theories of generalization and optimization in over-parameterized systems. He explores how interpolating models can be statistically optimal, how gradient descent converges in non-convex landscapes via the PL* condition, and how kernel methods can be enhanced to perform feature learning. ACM Fellow (2023) Editor-in-Chief, SIAM Journal on Mathematics of Data Science (2024–present) Advising and Grants: Belkin actively mentors students and collaborators such as Adityanarayanan Radhakrishnan , Daniel Beaglehole , and Chaoyue Liu , who are frequent co-authors. He is a Principal Investigator (PI) in the Collaboration on the Theoretical Foundations of Deep Learning , funded by the NSF and Simons Foundation. He is also an external collaborator with the Eric and Wendy Schmidt Center at the Broad Institute and part of the NSF-funded TILOS AI Institute . Laboratories and Teams: While not explicitly named, his research group at UCSD is deeply involved in theoretical machine learning, focusing on the intersection of statistics, optimization, and deep learning. His work often involves large-scale collaborations and is closely tied to initiatives like SIMODS and TILOS.

• Multimodal Human Behavior Modeling
• Health Informatics
• Affective Computing
• +7 more

Zakia Hammal is an Assistant Research Professor with dual appointments at Carnegie Mellon University, holding positions in the Robotics Institute within the School of Computer Science and the Department of Biomedical Engineering in the College of Engineering. Her work bridges computer science, machine learning, artificial intelligence, and social/behavioral psychology to advance computational models for human behavior analysis. Dr. Hammal's educational background includes a PhD in Computer Science, a Master of Artificial Intelligence and Algorithmic with specialization in Image Processing, and an Engineer's degree in Computer Science with specialization in Computer Systems. Her academic journey has positioned her at the intersection of technical expertise and healthcare applications. Her research focuses on multimodal human behavior modeling in social interaction, with particular emphasis on health informatics and affective computing (Emotion AI). Dr. Hammal's work has pioneered computational models for multimodal assessment of psychiatric disorders, including depression severity evaluation, automatic pain intensity measurement, assessment of expressiveness in children with facial abnormalities, analysis of non-verbal communication in mother-infant interaction, and identification of behavioral markers in autism spectrum disorder. Her approach integrates computer vision, machine learning, and behavioral psychology to create systems that can objectively measure human behaviors that are often subjective in clinical settings. Analysis of her recent publications reveals a consistent trajectory toward more sophisticated multimodal approaches to healthcare challenges, particularly in pain assessment and mental health diagnostics. Her work increasingly emphasizes interpretable AI models that can translate complex behavioral patterns into clinically meaningful insights, with growing attention to applications for vulnerable populations including infants, elderly patients, and those with craniofacial abnormalities or autism spectrum disorder. Women in AI Awards North America 2023 – AI Researcher of the Year Award Outstanding Reviewer Award at FG 2015 Best Paper award at ACII 2015 Outstanding Paper award at ICMI 2012 Dr. Hammal has secured significant research funding, primarily from the U.S. National Institutes of Health, including an R01 grant for developing a Multimodal Behavioral AI platform for pain assessment and management, and additional grants for automatic pain assessment in older adults with dementia. Her leadership extends to mentoring through her involvement in organizing workshops and conferences that train the next generation of researchers in affective computing and health informatics. As an active leader in her field, Dr. Hammal serves as ACM ICMI Steering Board Committee Member, Associate Editor for IEEE Transactions on Affective Computing and IEEE Transactions on Multimedia, and has organized numerous influential workshops including the International Workshop on Automated Assessment of Pain and Face and Gesture Analysis for Health Informatics. She is set to serve as Program Chair for FG 2025, ACII 2025, and ICMI 2026, demonstrating her growing influence in shaping the future direction of research in multimodal interaction and affective computing.

Muhammad Abdul-MageedView profile

Associate Professor

• Natural Language Processing
• Deep Learning
• Socio-Pragmatics
• +5 more

Dr. Muhammad Abdul-Mageed is an Associate Professor in the School of Information at The University of British Columbia, with joint appointments in Linguistics and an associate membership in Computer Science. He holds the Canada Research Chair in Natural Language Processing and Machine Learning. His research focuses on deep learning, socio-pragmatics, and speech/language technologies, particularly for Arabic and African languages. He leads the UBC Deep Learning & NLP Group and co-directs SSHRC-funded grants like I Trust AI and Ensuring Full Literacy. He is a founding member of the Center for Artificial Intelligence Decision making and Action and a member of the Institute for Computing, Information, and Cognitive Systems. His work spans automatic speech recognition, machine translation, computational socio-pragmatics, and low-resource language technologies. Notable projects include developing Arabic speech recognition systems, multidialectal Arabic benchmarks, and tools for African language processing. He has authored over 100 peer-reviewed papers and leads initiatives like the NADI Arabic Dialect Identification shared task and the NileChat project for culturally-aware LLMs. His research aims to create equitable, socially-aware AI systems for health, social media, and information management.

James GlassView profile

Research Fellow

• Machine Learning
• Natural Language Processing
• Speech Recognition
• +3 more

James Glass is a Senior Research Scientist at the Massachusetts Institute of Technology (MIT) and heads the Spoken Language Systems Group within MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL). He is also affiliated with the Harvard-MIT Division of Health Sciences and Technology. His research spans automatic speech recognition, multimodal learning, and spoken language understanding, with applications in healthcare and video analysis. Education: SM and PhD in Electrical Engineering and Computer Science from MIT His work focuses on paralinguistic speech analysis, health markers in speech, and the intersection of speech and natural language processing. Recent trends emphasize audio-visual alignment, recursive reasoning, and AI applications in cognitive disorder diagnosis. Scientific awards include IEEE Fellow, ISCA Fellow, and Associate Editor for IEEE Transactions on Pattern Analysis and Machine Intelligence. His group explores unsupervised learning, speaker verification, and social text analysis. James leads the Spoken Language Systems Group at CSAIL, collaborating with institutions like IBM and Harvard-MIT Division of Health Sciences and Technology. His research integrates vision-language models, neural audio codecs, and self-supervised frameworks.

• Natural Language Processing
• Natural Language Generation
• Machine Learning
• +4 more

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

Erkut ErdemView profile

Professor

• Computer Vision
• Machine Learning
• Visual Saliency Prediction
• +5 more

Erkut Erdem is a Professor in the Department of Computer Engineering at Hacettepe University, where he leads the Computer Vision Laboratory (HUCVL). His research focuses on computer vision and machine learning, particularly on incorporating different kinds of context (spatial, temporal and cross-modal) into visual processing across all levels from low to high-level vision. He received his Ph.D. (2008), M.Sc. (2003), and B.Sc. (2001) from Middle East Technical University. Prior to joining Hacettepe University in 2010, he completed a post-doctoral fellowship at Ecole Nationale Supérieure des Télécommunications (2009-2010) and held visiting researcher positions at UCLA (2007) and Virginia Tech (2004). His current research interests include Visual Saliency Prediction, Automatic Image Description, Video/Photoset Summarization, Image Filtering, and Image Editing. Recent work has focused on multimodal learning with video-language models, diffusion-based image editing, and event-based vision for low-light conditions. His research has been published in top venues including NeurIPS, ICLR, ICCV, SIGGRAPH, and ACL. He has received significant recognition including The Young Researcher Award from Turkish Academy of Sciences and being named a 2022 Outstanding Associate Editor of IEEE Transactions on Multimedia. He has secured multiple research projects funded by TUBITAK and received gift funds from Adobe Research for text-guided image synthesis work. Current Teaching: BBM202: Algorithms, AIN434/BBM444: Fundamentals of Computational Photography Graduate Supervision: 6 current Ph.D. students, numerous recent graduates including Burak Ercan (2024) and Aysun Kocak (2023) Professional Affiliations: Co-affiliated with Koç University and İş Bank AI Center (KUIS AI)

• Transportation Engineering
• Road Safety
• Remote Sensing
• +5 more

Dr. Karim El-Basyouny is a Killam Laureate Professor and City of Edmonton Urban Traffic Safety Research Chair at the University of Alberta's Faculty of Engineering, where he serves as Associate Dean (Research Infrastructure and Innovation) in the Civil and Environmental Engineering Department. A licensed Professional Engineer in Alberta, he holds advanced degrees in Transportation Engineering from the University of British Columbia and has dedicated his career to advancing road safety through data-driven management frameworks. His academic credentials include: Doctor of Philosophy, Civil Engineering, University of British Columbia, 2011 Engineering Management Sub-specialization, Civil Engineering, University of British Columbia, 2010 Master of Applied Science, Civil Engineering, University of British Columbia, 2006 Bachelor's degree (ABET Equivalent), Civil & Environmental Engineering, United Arab Emirates University, 2003 El-Basyouny's research pioneers the integration of remote sensing, machine learning, and statistical modeling to enhance transportation safety. His work develops automated tools for infrastructure digitization, collision prediction, and speed management, treating safety as a systemic product requiring management frameworks. Key contributions include LiDAR-based road feature extraction, network-level safety evaluations, and frameworks for vision-zero outcomes that address both human-driven and autonomous vehicle contexts. His recent publications demonstrate a cohesive research trajectory centered on leveraging point cloud data and computational intelligence for safety management. Over 15 major publications since 2021 focus on automated infrastructure assessment (light pole detection, clear zone mapping, vertical clearance evaluation), weather-impact modeling, and enforcement resource optimization. This body of work bridges transportation engineering with computer vision and operations research to create scalable safety solutions. His scientific contributions have been recognized with prestigious honors including: 2024 Killam Annual Professorship Award 2024 Road Safety Achievement Award from TAC 2023 Donald Stanley Award for environmental engineering 2022 Faculty of Engineering Graduate Teaching Award 2021 Daniel B. Fambro Student Paper Award As an academic leader, El-Basyouny actively mentors graduate students and secures significant research funding through his endowed chair position. He currently recruits fully-funded PhD and postdoctoral candidates specializing in remote sensing applications, machine learning, and geomatics for road digitization projects. His research group collaborates with national safety committees and municipal agencies to translate findings into policy, while he serves on editorial boards for Transportation Research Record and Analytic Methods in Accident Research. The research group operates at the intersection of transportation engineering and computational science, developing automated frameworks that merge sensor technologies with data processing tools. Current projects focus on semantic segmentation of 3D point clouds, safety implications of infrastructure digitization, and machine learning applications for road feature extraction in both urban and rural environments.

John C. DoyleView profile

Professor

• Control Theory
• Dynamical Systems
• Network Science
• +4 more

John C. Doyle is the Jean-Lou Chameau Professor of Control and Dynamical Systems, Electrical Engineering, and BioEngineering at the California Institute of Technology (Caltech), where he holds appointments in the Division of Engineering and Applied Science with primary affiliation in the Control and Dynamical Systems Department. His research bridges theoretical foundations with applications across biological, technological, medical, and ecological networks. He earned a BS and MS in Electrical Engineering from MIT (1977) and a PhD in Mathematics from UC Berkeley (1984), followed by consultancy at Honeywell Systems and Research Center (1976-1990). MIT: BS & MS in Electrical Engineering (1977) UC Berkeley: PhD in Mathematics (1984) Doyle's research centers on universal laws and architectures in complex systems, emphasizing robustness-efficiency tradeoffs, speed-accuracy tradeoffs (SATs), diversity-enabled sweet spots (DeSS), bowtie/hourglass structures, and evolvability. His work pioneers System Level Synthesis (SLS) for control systems with sparse, local, saturating, delayed, noisy, quantized, and distributed (SLSDNQD) components, integrating control theory, computation, communication, and machine learning to address challenges from neural networks to infrastructure resilience. Key concepts include virtualization, horizontal transfer, and virality in multiscale systems. Analysis of his publication trends reveals consistent interdisciplinary impact across neuroscience (brain connectivity modeling), systems biology (metabolic oscillations), network science (internet topology), and physics (turbulence, earthquakes), with recurring themes of robust-efficiency limits and architectural principles governing complex networks. His work demonstrates exceptional translation from abstract theory to practical tools like the Matlab Robust Control Toolbox and Systems Biology Markup Language (SBML). His scientific recognition includes: 1990 IEEE Baker Prize (ranked among top 10 most important mathematics papers 1981-1993) Three IEEE Automatic Control Transactions Awards (1998, 1999, 2021) ACM Sigcomm Paper Prize (2004) and Test of Time Award (2016) IEEE Control Systems Field Award (2004) Multiple early-career honors including IEEE Centennial Outstanding Young Engineer (1984) Doyle has mentored generations of students whose contributions include foundational software tools adopted globally. His research has secured sustained funding from NSF, NIH, and other agencies supporting theoretical advances in control frameworks and their applications to biomedical systems, network infrastructure, and environmental modeling. The SBML initiative exemplifies his group's impact in standardizing computational biology research. He leads a highly collaborative research ecosystem at Caltech that integrates engineers, biologists, neuroscientists, and computer scientists to develop universal principles for complex networks. Current efforts focus on translating theoretical insights into health technologies, resilient infrastructure, and climate-responsive systems through the application of robust-efficiency frameworks to emerging challenges in cyber-physical and biological domains.

Elsayed IssaView profile

Assistant Professor

• Natural Language Processing
• Second Language Acquisition
• Arabic Linguistics
• +6 more

Elsayed Issa is an Assistant Professor in the School of Languages and Cultures at Purdue University. Specializing in Computational Linguistics and Arabic, his interdisciplinary research bridges Natural Language Processing (NLP) with Second Language Acquisition (SLA), focusing on conversational AI and speech technology for under-resourced languages. Ph.D. in Linguistics from the University of Arizona (2023) Research integrates NLP, conversational AI, and SLA methodologies Develops tools for computer-assisted pronunciation training (CAPT) Focuses on Arabic dialectology and large language models (LLMs) His work employs Transformer architectures and end-to-end machine learning to enhance language learning systems. Recent projects include ArabiBot development and dialect identification models. He specializes in speech-to-text systems , prosody modeling , and emotional speech analysis for Arabic language learning applications.

Karen LivescuView profile

Professor

• Speech Processing
• Language Processing
• Machine Learning
• +5 more

Karen Livescu is a Professor at the Toyota Technological Institute at Chicago (TTIC), a philanthropically endowed graduate institute for computer science located on the University of Chicago campus. She also serves as a courtesy faculty member in the Department of Computer Science at the University of Chicago and is an Affiliated Scholar at the Data Science Institute there. Her research focuses on advancing speech and language processing through innovative machine learning approaches. Education: PhD in Electrical Engineering and Computer Science from MIT (2005) S.M. from MIT Department of Electrical Engineering and Computer Science (1999) A.B. in Physics from Princeton University (1996) Karen's research spans multiple dimensions of speech and language processing with particular emphasis on speech recognition, spoken language understanding, and multimodal processing. She has made significant contributions to articulatory feature-based speech recognition, self-supervised learning for speech representation, and sign language processing. Her work consistently bridges machine learning techniques with linguistic and speech science knowledge, focusing on creating more robust, interpretable, and inclusive speech processing systems that can handle diverse languages and modalities. Her recent publication trajectory reveals a strong focus on self-supervised learning for speech representation, multilingual speech processing, and sign language understanding. She has been instrumental in developing benchmark frameworks like SUPERB and ML-SUPERB that have become standard evaluation tools in the speech community. Her work increasingly addresses critical challenges in low-resource language scenarios, language disparities in speech technology, and ethical considerations in real-world deployment. Scientific Awards: Best Paper award at EMNLP 2024 for 'Towards robust speech representation learning for thousands of languages' Best Student Paper Award at ASRU 2023 Best Short Paper Award at CRAC 2021 Top system at WMT-SLT 2023 Karen has successfully advised numerous PhD students and postdoctoral researchers who have gone on to faculty positions at institutions like University of Waterloo, University of Edinburgh, and Stellenbosch University, as well as industry roles at major technology companies including Google, Meta, and NVIDIA. Her research group has secured significant funding for projects including the development of the SLUE benchmark for spoken language understanding and the SUPERB framework for evaluating self-supervised speech models. She has been actively involved in organizing workshops and symposia that bring together researchers in speech and language processing. Karen leads the Speech and Language at TTIC (SL@TTIC) research group, which maintains a strong collaborative relationship with researchers at the University of Chicago and other institutions. The group has been particularly active in advancing sign language processing through projects like ChicagoFSWild and OpenASL, while also making significant contributions to spoken language understanding and multilingual speech recognition. Her team regularly participates in community challenges and benchmarks, helping to push the field forward through open science and collaborative evaluation frameworks.

Mohsen LesaniView profile

Associate Professor

• Reliability of software systems
• Security of software systems
• Concurrent systems
• +12 more

Mohsen Lesani is an Associate Professor in the Computer Science and Engineering Department at the University of California, Santa Cruz's Baskin School of Engineering. His research focuses on reliability and security of software systems, particularly concurrent and distributed systems, with recent emphasis on secure replicated systems and distributed machine learning. Dr. Lesani received his PhD from UCLA, MS in artificial intelligence from Sharif University of Technology, and BS in software engineering from University of Tehran. He was previously a postdoc at MIT. His educational background provides a strong foundation for his interdisciplinary research spanning programming languages, distributed systems, and security. His research interests center on creating reliable and secure distributed systems. Current projects include resilient and secure distributed systems, heterogeneous and reconfigurable secure distributed systems, automatic analysis and synthesis of replicated objects, verification of distributed systems, data analytics, secure exchange across blockchains, machine learning for performance models, domain-specific languages and type systems, and automatic fence insertion for concurrent systems. His work bridges theoretical foundations with practical implementations to address real-world challenges in distributed computing. Lesani's research has been recognized with several prestigious awards including the NSF CAREER award in 2020 and DARPA YFA award in 2022. His work has also received the SIGPLAN Research Highlight in 2019, a distinguished paper award at OOPSLA 2018, and a best paper award at ISSRE 2015. These accolades reflect the impact and quality of his contributions to the field. He actively mentors PhD students in the Safe and Secure Software (S3) lab, including Xiao Li, Eric Chan, Javad Saber-Latibari, and Tejas Mane. His research has been supported by multiple NSF grants, demonstrating sustained funding for his innovative work. Lesani serves on program committees for major conferences including POPL, PLDI, OOPSLA, and DISC, contributing to the academic community. Lesani leads the Safe and Secure Software (S3) lab at UC Santa Cruz, where his team works on cutting-edge research in distributed systems, programming languages, and security. The lab fosters a collaborative environment where theoretical insights are translated into practical systems that address real-world challenges in reliability and security of distributed applications.

• Natural Language Processing
• Computational Linguistics
• Machine Learning
• +7 more

Barbara Plank is a full professor and chair for AI and Computational Linguistics at Ludwig Maximilian University of Munich (LMU), where she heads the Munich AI and NLP (MaiNLP) lab and co-directs the Center for Information and Language Processing (CIS). She additionally serves as a visiting full professor at the IT University of Copenhagen, maintaining active dual institutional affiliations in computational linguistics and NLP research. Her research focuses on human-centric natural language processing challenges, particularly learning under sample selection bias (domain adaptation, transfer learning) and annotation bias, learning with limited data through continual/semi-supervised/weakly-supervised methods, multimodal learning at language-vision-speech interfaces, and fortuitous supervision for variety-space aware language understanding. She pioneers methodologies addressing human label variation as a critical factor in model robustness rather than mere noise. Recent publications (2024-2025) reveal dominant trends in modeling human label variation across NLP tasks, especially natural language inference and entity recognition, alongside dialectal language processing and LLM evaluation frameworks. Her work systematically investigates how human disagreement in annotations can be leveraged to build more robust, adaptable systems rather than treated as errors. Scientific recognition includes: ERC Consolidator Grant for the DIALECT project advancing natural language understanding for non-standard languages and dialects ACL 2024 Area Chair Award for the paper 'VariErr NLI: Separating Annotation Error from Human Label Variation' Leading the MaiNLP lab at CIS (LMU), she directs research integrated with MCML (Munich Center for Machine Learning), Munich Intelligent Robotics, ELLIS Unit Munich, UniDive, and COST action. Current projects include ERC-funded DIALECT and KLIMA-MEMES, focusing on human-facing NLP solutions for real-world language diversity challenges. She actively shapes the field through ACL leadership as VP-Elect and numerous keynotes emphasizing human-centric approaches. The MaiNLP lab at Akademiestr. 7, 80799 Munich, drives innovation in computational linguistics through interdisciplinary collaboration, maintaining strong ties with European research networks while developing practical applications for language variation and robust NLP systems. The lab's work directly informs her teaching in LMU's Computational Linguistics programs, bridging research and education in cutting-edge NLP methodologies.