Colin Raffel , currently an Associate Professor at the University of Toronto and Associate Research Director at the Vector Institute , is a leading researcher in machine learning and natural language processing . His career spans roles at Hugging Face (Faculty Researcher), Google Brain (Senior Research Scientist), and UNC Chapel Hill (Assistant Professor). Education: PhD in Electrical Engineering (Columbia), MA in Music/Science (Stanford), BA in Mathematics (Oberlin) Key affiliations: Google Brain (2016-2020), Hugging Face (2021-present), Vector Institute (2023-present) His research focuses on language model development , attention mechanisms , efficient machine learning , and music information retrieval . Recent work explores model merging , parameter-efficient fine-tuning , and data-constrained language models . Teaching : Has instructed courses at University of Toronto and UNC Chapel Hill on Neural Networks , Deep Learning , and Information Theory . Academic service includes organizing ICLR workshops and serving as Senior Area Chair for NeurIPS and EMNLP . Notable awards : NSF CAREER (2022), Caspar Bowden Award (2023), NeurIPS Outstanding Paper (2023) Key contributions : Core developer of WT5 , Git-Theta , and mir_eval software
Furong Huang is an Associate Professor at the University of Maryland's Department of Computer Science, with affiliations at the Institute for Advanced Computer Studies, Center for Machine Learning, Maryland Robotics Center, and Applied Mathematics, Statistics, and Scientific Computation Program. Her research bridges trustworthy machine learning, sequential decision-making, and foundation models for robotics, emphasizing reliability, interpretability, and ethical standards. Research Interests: Trustworthy AI Generative AI Reinforcement Learning AI Security Algorithmic Fairness Foundation Models for Robotics Recent Publications span leading conferences (NeurIPS, ICML, ICLR, CVPR) and journals, focusing on: Robustness in Vision-Language Systems Trustworthy Generative AI Foundation Models for Sequential Decision-Making AI Security and Watermarking Scientific Awards MIT TR35 Innovator Under 35 (Asia Pacific 2022) Best Paper Award, AdvML Frontier Workshop, NeurIPS 2024 NSF NAIRR Pilot Awardee Microsoft Accelerate Foundation Models Research Award (2023) JP Morgan Faculty Research Awards (2019–2022) Advising and Grants : Her lab has graduated students to roles at OpenAI, Google, Meta, and Netflix. Research funded by DARPA, NSF, ONR, AFOSR, and industry partners like Microsoft, Adobe, and Capital One. Labs & Teams : Leads research groups focused on Trustworthy AI and Robotics at the University of Maryland, collaborating with the Maryland Robotics Center and Applied Mathematics Program.
Huan Zhang serves as an Assistant Professor in the Department of Electrical and Computer Engineering at the University of Illinois Urbana-Champaign (UIUC), with affiliate appointments in the Department of Computer Science and the Coordinated Science Laboratory. His research focuses on building trustworthy AI systems with formal verification techniques to provide provable guarantees for safety-critical applications, particularly in machine learning and neural networks. Dr. Zhang received his Ph.D. in Computer Science from UCLA in 2020, advised by Professor Cho-Jui Hsieh. His academic journey includes an M.S. in Computer Engineering from UC Davis (2014) and a Bachelor of Engineering from Zhejiang University (2012). Prior to joining UIUC, he completed a postdoctoral fellowship at Carnegie Mellon University (2021-2023) with Professor Zico Kolter. Huan Zhang's research program centers on formal verification of machine learning systems, with particular emphasis on neural network verification, AI safety, robustness, and reliability. He pioneered the linear bound propagation-based verification framework that enables formal verification for networks with millions of neurons. His work spans five major research categories: formal verification of machine learning, training trustworthy ML models, machine learning safety and adversarial attacks, reinforcement learning safety, and optimization for scalable machine learning. His CROWN framework (NeurIPS 2018) established a foundational approach for neural network verification through efficient linear bound propagation. His recent publications demonstrate a strategic expansion from foundational verification techniques toward increasingly complex systems including large language models, vision-language models, and robotic control systems. The research trajectory shows a clear progression from theoretical frameworks to practical implementations with real-world applications, particularly in safety-critical domains. His work increasingly bridges formal methods with practical AI deployment requirements. Winner of International Verification of Neural Networks Competition (VNN-COMP) as team leader (2021-2024) Schmidt Futures AI2050 Early Career Fellowship ($300,000 research grant) Adversarial Machine Learning (AdvML) Rising Star Award (2021) IBM PhD Fellowship (2018) Dr. Zhang leads the development of α,β-CROWN, a neural network verifier that has won VNN-COMP 2021-2023, and auto_LiRPA, a PyTorch-based library for perturbation analysis on general computational graphs. He has mentored numerous graduate students from CMU, UCLA, UIUC, and Columbia University. His research is supported by significant funding including the Schmidt Futures fellowship and industry collaborations. He teaches courses including ECE 120, ECE 484, ECE 584, and ECE 598 HZ on topics ranging from computing fundamentals to safe autonomy and machine learning. Dr. Zhang maintains active research collaborations across multiple institutions and is affiliated with UIUC's Coordinated Science Laboratory. His work has significant implications for safety-critical AI applications in autonomous systems, healthcare, and other mission-critical domains where reliability guarantees are essential. He regularly gives guest lectures at institutions including Yale, Stony Brook, and the University of Nebraska Lincoln on formal verification techniques.
Prof. Luke Zettlemoyer is an Adjunct Professor of Computer Science and Engineering at the University of Washington, with affiliations to the Department of Linguistics. He focuses on machine learning, natural language processing, and multimodal systems, contributing to advancements in large language models, ethical AI, and scalable architectures. His research addresses challenges in model alignment, generalization, and cross-domain integration. Key research interests include multimodal reward models, efficient tokenization strategies, and model optimization techniques. He has explored topics such as neural trajectories for robot learning, content-adaptive image processing, and ethical mitigation of verbatim data reproduction. His publications span 2023–2025, emphasizing practical applications of AI in robotics, vision-language systems, and scalable retrieval-based models. While no formal awards are listed, his work reflects significant contributions to foundational AI research.
Arman Cohan is an Assistant Professor in the Department of Computer Science at Yale University, where he leads the Yale NLP Lab since its founding in January 2023. His research spans natural language processing and machine learning with emphasis on language modeling, representation learning, retrieval systems, and specialized domain applications including scientific discovery and AI for science. His primary research interests include: Natural Language Processing Machine Learning Large Language Models Information Retrieval AI for Science Scientific Problem-Solving Recent publications (2025) demonstrate intense focus on evaluating and advancing LLM capabilities across multimodal reasoning, scientific claim verification, financial domain applications, and biological modeling. The lab consistently produces high-impact work accepted at top-tier conferences including ACL, EMNLP, and ICLR, with 11 papers at ACL 2025 alone. Scientific awards include: Best Paper Award at AI4Research Workshop (IJCAI 2024) Outstanding Paper Award at EACL 2023 Best Paper Award at ACL 2024 for Olmo language model research Professor Cohan actively advises PhD students including Kaili Liu, Jacob Dunefsky, Alan Li, Yilun Zhao, and has graduated researchers such as Linyong Nan (now at Zoom) and Ansong Ni (now at Meta). The lab maintains strong industry partnerships and receives substantial research funding as evidenced by its prolific output and conference presence. The Yale NLP Lab hosts the annual New England NLP Workshop and regularly features speakers from leading institutions including Meta AI, Allen Institute for AI, and DeepMind, fostering a collaborative environment for advancing NLP research.
Vineeth N Balasubramanian is a Professor in the Department of Computer Science & Engineering at the Indian Institute of Technology Hyderabad, with affiliate faculty status in the Department of Artificial Intelligence. His research focuses on the intersection of deep learning, machine learning, and computer vision, emphasizing explainability, robustness, and real-world applications. He leads Lab 1055, which investigates problems such as Explainable and robust AI/ML systems Lifelong learning in evolving environments Multimodal vision-language models Applications in agriculture, autonomous navigation, and human behavior analysis His recent work includes causal reasoning in transformers, vision-language model capabilities, and drone-based object detection. Funded by organizations like Google, Microsoft, Intel, and DST, he has received multiple awards including the World's Top 2% Scientists (2022-23), INSA/INAE Fellowships, and Best Paper recognitions. Lab 1055 collaborates with institutions like CMU, UBC, and Monash University, contributing to cutting-edge advancements in AI.
Chris Donahue is an Assistant Professor in the Computer Science Department at Carnegie Mellon University . He also serves as a part-time Research Scientist at Google DeepMind on the Magenta team. His work focuses on leveraging generative AI to enhance human creativity, particularly in music. Education: PhD in Computer Science (UC San Diego), Postdoctoral Scholar (Stanford University) His research spans controllable generative modeling of music and audio , with a focus on real-time interactive systems. Projects like Piano Genie , Beat Sage , and Copilot Arena demonstrate his commitment to real-world deployment. His Generative Creativity Lab (G-CLef) explores AI applications beyond music, including programming and natural language. Recent publications highlight advancements in multimodal music evaluation , real-time adaptation , and AI-driven sound morphing . He co-developed Magenta RealTime , an open-weight real-time music generation model, and MusicFX DJ Mode . Scientific Awards: Best Paper Award (top 1) at NAACL Student Research Workshop 2025 Best Paper Award (top 1% of submissions) at CHI 2025 Best Paper Runner-up at ISMIR 2021 He co-advises PhD students like Wayne Chi (NDSEG Fellow) and mentors Irmak Bukey . His lab receives support from the AIxArts incubator fund at CMU .
Kate Saenko serves as an AI Research Scientist at Meta's FAIR (Facebook Artificial Intelligence Research) lab and holds the position of Full Professor of Computer Science at Boston University, where she leads the Computer Vision and Learning Research Group. Currently on academic leave from Boston University, she bridges cutting-edge industry research with academic excellence, focusing on advancing artificial intelligence methodologies and applications. Her educational background includes a PhD in Electrical Engineering and Computer Science (EECS) from the Massachusetts Institute of Technology (MIT), followed by postdoctoral training at the University of California, Berkeley and Harvard University. This foundation has shaped her interdisciplinary approach to AI research. Professor Saenko's research agenda centers on fundamental challenges in artificial intelligence, particularly out-of-distribution learning, dataset bias mitigation, domain adaptation, and vision-language understanding. Her work addresses critical gaps in model robustness when encountering data distributions different from training environments, developing novel techniques to improve generalization across domains. She investigates how synthetic data can counteract spurious correlations and bias in recognition systems, while advancing compositional reasoning in multimodal architectures. Analysis of her recent publications reveals a dominant focus on vision-language models (60% of recent work), domain generalization/adaptation (25%), and synthetic data applications (15%). Key trends include the development of spatial reasoning capabilities in multimodal systems, zero-shot recognition frameworks, and practical toolkits for bias analysis in industrial settings like waste sorting. Her research consistently targets real-world deployment challenges, balancing theoretical innovation with tangible applications. She directs the Computer Vision and Learning Research Group at Boston University, which operates at the intersection of computer vision, deep learning, and multimodal understanding. The group maintains strong industry collaborations through Meta's FAIR and previously engaged with the MIT-IBM Watson AI Lab. Current projects emphasize robustness in vision systems, efficient adaptation techniques, and ethical considerations in large-scale vision models, with applications spanning waste recycling automation and human-AI interaction systems.
Lerrel Pinto is an Assistant Professor of Computer Science at the Courant Institute of Mathematical Sciences at New York University (NYU), where he leads the General-purpose Robotics and AI Lab (GRAIL) as part of the CILVR research group. His work bridges the gap between theoretical machine learning and practical robotics applications, with a focus on enabling robots to generalize and adapt in real-world environments. Dr. Pinto received his undergraduate degree from IIT Guwahati, followed by a PhD from the Robotics Institute at Carnegie Mellon University (CMU). He then completed a postdoctoral fellowship at the University of California, Berkeley before joining NYU as faculty. His research program centers on robot learning and decision making, with several key thrusts that demonstrate his innovative approach to robotics. Pinto's work emphasizes large-scale learning techniques that leverage both extensive data and sophisticated model architectures. A significant portion of his research focuses on representation learning for sensory data, particularly developing methods that enable robots to make sense of visual, tactile, and auditory inputs. His lab has made notable contributions to reinforcement learning algorithms that allow robots to adapt to new scenarios with minimal retraining. Pinto also champions open-source robotics , developing affordable robot platforms that democratize access to robotics research. Analysis of Pinto's recent publications reveals a strong trend toward multimodal perception in robotics, integrating visual, tactile, and auditory information to create more robust robot systems. His work increasingly focuses on zero-shot and few-shot learning capabilities, enabling robots to handle novel situations without extensive retraining. There's also a clear progression toward general-purpose robotics , moving away from task-specific solutions toward more flexible systems that can handle diverse real-world challenges. Dr. Pinto's scientific contributions have been recognized with several prestigious awards: Sloan Research Fellowship (2025) NSF CAREER Award (2024) RAL Early Career Award (2024) Best Student Paper Award at ICRA (2016) Outstanding Paper Award at MFM-EAI workshop at ICML (2024) Best Paper Award at NGSM workshop at ICML (2024) Best Student Paper Award at RSS (2023) As an advisor, Pinto has mentored numerous students who have gone on to impactful careers in both academia and industry. His former PhD student Denis Yarats co-founded Perplexity.AI, while Mahi Shafiullah became a postdoc at UC Berkeley and Meta AI. Many of his Masters students have pursued PhDs at top institutions like CMU, MIT, and Stanford, or joined leading robotics companies including 1X, Fauna Robotics, and NVIDIA. Pinto's lab has secured significant research funding, including the NSF CAREER award and likely other grants supporting his robotics research program. The General-purpose Robotics and AI Lab (GRAIL) that Pinto leads brings together a diverse team of researchers working on cutting-edge robotics challenges. The lab maintains strong collaborations with industry partners and other academic institutions, facilitating technology transfer and real-world impact. GRAIL's research spans multiple robotics platforms and focuses on developing algorithms that enable robots to learn from diverse experiences and generalize across environments.
Heng Huang is the Brendan Iribe Endowed Professor in the Department of Computer Science and the Department of Electrical and Computer Engineering at the University of Maryland, College Park . He earned his Ph.D. in Computer Science from Dartmouth College and holds prior degrees from Shanghai Jiao Tong University. His research focuses on advancing the foundations and applications of artificial intelligence, particularly in machine learning, data mining, natural language processing, computer vision, and biomedical informatics . His work integrates large-scale optimization, fairness, and robustness in deep learning systems. Heng Huang’s recent publications demonstrate a strong trend in large language models, federated learning, model watermarking, continual learning, and medical image analysis . His work appears consistently in top venues like NeurIPS, ICML, CVPR, ICLR, and MICCAI, reflecting a broad impact across theoretical and applied AI. He actively mentors students and postdocs, seeking highly motivated researchers in machine learning and related domains. His work has significant implications for healthcare, privacy, and trustworthy AI.
Walid Magdy is a Professor at the School of Informatics, The University of Edinburgh, where he is a faculty member at the Institute for Language, Cognition and Computation (ILCC). He holds a PhD from Dublin City University's School of Computing and has been a faculty fellow at The Alan Turing Institute. With substantial industry experience, he has worked for Qatar Computing Research Institute (QCRI), Microsoft, and IBM, where he filed nine patents. His research spans computational social science and natural language processing. In computational social science, he focuses on social content analysis, political bias detection, polarization, and users' behavior analysis/prediction, with particular attention to understudied communities. His natural language processing work centers on social applications including sentiment analysis, sarcasm detection, and hate-speech identification, with specialized expertise in Arabic and its dialects as well as Persian. Professor Magdy's recent publications demonstrate a consistent focus on Arabic language processing and social media analysis, with growing attention to cultural factors in NLP systems. His work bridges computational methods with social science questions, particularly regarding misinformation, cultural differences in online behavior, and the impact of platform policies on diverse linguistic communities. He has expanded his research to include Persian language processing and continues to explore social dynamics across different cultural contexts. His significant contributions have been recognized with multiple awards: Best Demo Award for TwiXplorer at CSCW 2024 Outstanding Paper Award at ACL 2024 Best Paper Award Runner-up at SocInfo 2022 Best Paper Award at WANLP-EMNLP 2022 Best Paper Award at AIRS 2015 Best Dataset Award at ICWSM 2013 Professor Magdy serves as an editor for Springer's Social Network Analysis and Mining journal and has held leadership roles in major conferences including ACL 2023, ArabicNLP 2023, ICWSM 2022, and ASONAM 2022. He has co-organized numerous shared tasks focused on Arabic language processing and social media analysis, contributing significantly to community resources and benchmarks. He founded and directs the Social Media Analysis and Support for Humanity (SMASH) research group at the University of Edinburgh, which has organized the Summer School in Computational Social Science (SICSS-Edinburgh) in 2022, 2023, and 2025, training researchers in computational social science methodologies.
Professor Anna Korhonen is a leading academic at the University of Cambridge, holding positions as Professor of Natural Language Processing, Co-Director of the Language Technology Laboratory (LTL), Director of the Centre for Human-Inspired Artificial Intelligence (CHIA), and Fellow of Churchill College. Her work bridges computational linguistics, artificial intelligence, and interdisciplinary applications. Her research focuses on human-centric NLP with core interests in multilingual/low-resource systems, conversational AI, and responsible technology development. She emphasizes applications for social and global good, including healthcare, climate science, and equitable language technologies. Her methodology integrates cognitive modeling with machine learning to create interpretable, fair NLP systems. Key projects include ERC-funded initiatives like MultiConvAI (multilingual conversational AI) and Towards Globally Equitable Language Technologies , alongside Innovate UK's ESG RoboFactory and EPSRC's Modeling Idiomaticity project. Her work spans biomedical text mining ( PheneBank , LION ), educational technology ( EF Education First Research Lab ), and cross-lingual transfer learning. Fellow of the Association for Computational Linguistics (ACL) Fellow of ELLIS (European Laboratory for Learning and Intelligent Systems) Royal Society University Research Fellow (2005-2014) Google Faculty Award recipient EACL Chair Elect Korhonen actively supervises PhD/MPhil students in Computation, Cognition and Language programs and leads interdisciplinary collaborations across computer science, linguistics, and domain sciences. Her lab (LTL) develops foundational NLP techniques while addressing real-world challenges through partnerships with healthcare, environmental science, and education sectors. Current strategic initiatives include the Institute for Technology and Humanity and CHIA's human-inspired AI framework.
Adriana Kovashka is an Associate Professor at the University of Pittsburgh , affiliated with the School of Computing and Information and serving as Department Chair . Her academic journey began with BA degrees in Computer Science and Media Studies from Pomona College (2008) and a PhD in Computer Science from The University of Texas at Austin (2014). Joined Pitt’s faculty in January 2015 NSF CAREER awardee (2021) Google Faculty Research Award recipient Dr. Kovashka’s research spans Computer Vision , Machine Learning , and Natural Language Processing , focusing on visual rhetoric, weak multimodal supervision, and domain adaptation. She pioneered techniques for analyzing political imagery, developing robust object detection frameworks, and exploring the intersection of visual and textual persuasion through large-scale annotated datasets. Her recent work emphasizes geographic diversity in vision-language systems, audio-visual fusion for domain generalization, and shape-texture bias mitigation in CNNs. Key publications include groundbreaking studies on symbolic reasoning, multimodal dialogue systems, and ethical AI applications in education. Scientific honors include: NSF CRII Award (2016) NSF CAREER Award (2021) Pitt CRDF Award (2016, 2018) Best Paper at ECV Workshop (2021) Google Faculty Research Award (2016, 2018) Dr. Kovashka actively mentors students in multimodal learning projects and collaborates with interdisciplinary teams on NSF-funded initiatives. She co-organizes workshops like the first CVPR workshop on advertisement understanding and leads research groups exploring human-AI co-learning systems.
Professor Bernd Möbius is a leading academic in Phonetics and Phonology at the Department of Language Science and Technology, Saarland University. His research bridges phonetic theory with speech technology applications, focusing on text-to-speech systems, prosody modeling, and computational simulations of speech processes. Current research projects: DFG SFB 1102, C1: Information density and phonetic structure predictability DFG SFB 1102, C4: Slavic intercomprehension and surprisal theory (INCOMSLAV) Research Themes: Key areas include text-to-speech synthesis, speech prosody analysis, experimental methods in speech production/perception, information density in phonetics, and cross-linguistic studies of Slavic-Germanic languages. Scientific Contributions: Recent work explores Parkinson-induced dysarthria detection, breath noise acoustics, surprisal-driven speech behaviors, multilingual BERT models for idiomaticity, and perceptual consequences of acoustic adjustments.
Daniel Epstein is an Associate Professor in the Department of Informatics at the University of California, Irvine (UCI), where he leads the Personal Informatics Everyday (PIE) Lab. He holds a courtesy appointment in the Department of Computer Science and is affiliated with the Connected Learning Lab, Institute for Future Health, and Accessibility Research Collective. Epstein earned his Ph.D. in Computer Science & Engineering from the University of Washington and a B.S. in Computer Science from the University of Virginia. His research focuses on how personal tracking technology can better serve diverse users' needs, emphasizing ethical design, health equity, and long-term user engagement. Notable areas include pregnancy tracking, wearable technology for children and ADHD management, and AI-driven health chatbots. Epstein has received grants from NSF, Snap Inc., and UCI programs, totaling over $650k in funding. Epstein's work bridges HCI and health informatics, with over 60 peer-reviewed publications. He advises six Ph.D. students and has mentored numerous undergraduates. Awards include the ACM Senior Member distinction (2025), CHI Best Paper (2023), and NSF CAREER Award (2023). Professional service includes program committees for CHI, CSCW, and UbiComp, plus leadership roles in workshops like the Workgroup on Interactive Systems in Healthcare (WISH). Beyond academia, Epstein enjoys hiking, collecting turtle figures, and advocating for accessible technology.
