جستجوی استاد راهنمای دانشگاهی

Andrea Cavallaroمشاهده پروفایل

استاد

• Machine Learning
• Artificial Intelligence
• Computer Vision
• +۷ مورد دیگر

Andrea Cavallaro is a Full Professor at the École Polytechnique Fédérale de Lausanne (EPFL) and Director of the Idiap Research Institute. He holds dual appointments in the School of Engineering (STI) within the Institute of Electrical Engineering and Measurements (IEM) and the School of Engineering's Education Unit (SEL-ENS). His research focuses on machine learning for multimodal perception, privacy-preserving AI, and autonomous systems. Cavallaro earned his PhD in Electrical Engineering from EPFL in 2002 and has held leadership roles including Director of Research at Queen Mary University of London and Turing Fellow at The Alan Turing Institute. Education: PhD in Electrical Engineering (EPFL, 2002) Leadership: Idiap Director, Affiliate at ELLIS Society Editorial Roles: Editor-in-Chief of Signal Processing: Image Communication (2020–2023), Senior Area Editor for IEEE Transactions on Image Processing Research Interests: Machine learning for audio-visual sensing, privacy in AI, autonomous systems perception, and ethical AI frameworks. Key projects include AlignAI (trustworthy AI alignment) and CORSMAL (multimodal object manipulation). Recent articles explore privacy-aware AI models, adversarial attacks, and multimodal perception systems. His work bridges theoretical advancements with practical applications in robotics, healthcare, and education. Awards include the Royal Academy of Engineering Teaching Prize and IAPR Fellowship. Teaching: Leads courses on deep learning ethics and multimodal AI at EPFL. Advising: Supervises 11 PhD students in areas like privacy-preserving algorithms and autonomous systems. Labs/Teams: Coordinates Idiap’s Audiovisual Intelligence and Learning Lab (LIDIAP) and collaborates on projects like GraphNEx (explainable AI via graph neural networks).

Tiancheng Zhaoمشاهده پروفایل

پژوهشگر

• Multimodal Foundation Models
• AI Agents
• Vision-Language Models
• +۳ مورد دیگر

Tiancheng Zhao is a principal researcher at the Binjiang Institute of Zhejiang University and founder of the Om Artificial Intelligence Laboratory (Om AI Lab), dedicated to frontier open multimodal AGI research for building next-generation agents that transform work and life through advanced human-machine interaction. His academic credentials include: Ph.D. in Computer Science from Carnegie Mellon University (2016-2019) under Prof. Maxine Eskenazi, Prof. Louis-Philippe Morency, Prof. William W. Cohen, and Dr. Dilek Hakkani-Tur, with pioneering dissertation “Learning to Converse With Latent Actions” in end-to-end generative conversational models M.S. in Computer Science from Carnegie Mellon University (2014-2016) B.S. in Electrical Engineering from UCLA (2010-2014) with Summa Cum Laude, focusing on speech signal processing under Prof. Abeer Alwan Dr. Zhao’s research centers on multimodal foundation models and agents, tackling three core challenges: Multimodal Models for cross-modal representation learning in high-dimensional data, Learning to Learn for effective skill acquisition from diverse signals (supervised labels, rewards, meta-learning), and AI Agents for open-world understanding and complex decision-making. His work bridges computer vision, natural language processing, and real-world applications including healthcare analytics and remote sensing. Analysis of his 50+ publications reveals accelerating innovation in multimodal large language models (2024-2025), with emphasis on stable vision-language architectures (VLM-R1), agent orchestration frameworks, and domain-specific applications in geospatial analysis and healthcare. Key trends include solving long-tail distribution challenges in satellite imagery, developing human-like zooming capabilities for multimodal LLMs, and creating unified benchmarks for autonomous GUI testing. His scientific recognition includes: National Breakthrough Technology Award by Ministry of Science and Technology (2021) Microsoft Research Best & Brightest PhD (2018) BEST PAPER AWARD at SIGDIAL 2018 Best Paper Nomination at SIGDIAL 2016 Top 1 Outstanding Bachelor of Science Award at UCLA (2014) As Om AI Lab founder, Dr. Zhao leads research teams developing computational building blocks for human-AI collaboration. While specific student mentorship details aren’t public, his extensive publication record with junior co-authors indicates active research supervision. Current projects focus on practical system implementations for real-world multimodal agent deployment across diverse domains.

Shinji Watanabeمشاهده پروفایل

دانشیار

• Automatic Speech Recognition
• Speech Enhancement
• Spoken Language Understanding
• +۳ مورد دیگر

Shinji Watanabe is an Associate Professor at Carnegie Mellon University's Language Technologies Institute and a Courtesy Professor in the Electrical and Computer Engineering department. He holds a Ph.D. (Dr. Eng.) from Waseda University, Japan, and has held research roles at NTT Communication Science Laboratories, Mitsubishi Electric Research Laboratories (MERL), and Johns Hopkins University. His research focuses on automatic speech recognition, speech enhancement, and machine learning for speech processing. Watanabe has published over 300 peer-reviewed papers and received the Best Paper Award at IEEE ASRU 2019. His work emphasizes robust speech processing in challenging environments, multilingual models, and neural audio codecs. He leads the ESPnet toolkit development for end-to-end speech processing systems and contributes to technical committees like IEEE SLTC and APSIPA SLA. Recent research trends include streaming speech systems, universal speech enhancement (URGENT challenges), and fusion of discrete speech units with self-supervised representations. He explores scalable speech foundation models through benchmarks like ML-SUPERB 2.0 and investigates cross-modal audio-visual processing in challenges like MISP 2025. Education : B.S., M.S., Ph.D. (Waseda University) Affiliations : CMU Language Technologies Institute, CMU ECE, Former roles at MERL and Johns Hopkins Key Projects : ESPnet, OpenWhisper-Style Models, URGENT Challenge Frameworks

Chuang Ganمشاهده پروفایل

استادیار

• Artificial Intelligence
• Robotics
• Computer Vision
• +۷ مورد دیگر

Chuang Gan is an Assistant Professor at the University of Massachusetts Amherst, affiliated with the College of Information and Computer Sciences and the Department of Computer Science. His work focuses on advancing artificial intelligence, robotics, computer vision, and embodied agents through interdisciplinary research combining neural networks, physical simulations, and multimodal learning. Research interests include generative models, reinforcement learning, vision-language integration, and scalable autonomous systems. He explores topics like world modeling for robots, adaptive policy learning, and physics-driven AI. His projects often involve creating systems that learn from visual, auditory, and tactile inputs to perform complex tasks such as object manipulation, navigation, and decision-making in dynamic environments. Recent research trends emphasize embodied AI systems capable of long-horizon planning, compositional reasoning, and efficient learning from limited data. His work bridges theory and practice, with applications in robotics, simulation platforms, and multimodal generation. Key contributions include frameworks for 3D scene understanding, adaptive world models, and novel training paradigms for large language models. His research has been applied to robotics platforms like RoboDreamer and UBSoft, focusing on unbounded soft environments. Collaborations involve designing benchmarks for physical scene understanding (e.g., Physion++), and creating tools like DiffTactile for tactile simulation. His work often integrates principles from differential geometry, PDE dynamics, and game theory. Chuang Gan’s research group develops open-source tools and benchmarks, such as the SoftZoo robot co-design platform and the SOK-Bench situated reasoning benchmark. His team emphasizes scalable alignment methods beyond human supervision and explores ethical AI through principles like symmetry-enhanced training.

Matthieu Cordمشاهده پروفایل

استاد

• Computer Vision
• Machine Learning
• Deep Learning
• +۷ مورد دیگر

Matthieu Cord is a Professor at Sorbonne University and Scientific Director of valeo.ai, leading research in computer vision, deep learning, and computational cooking. He heads the MLIA team at ISIR Lab, focusing on multimodal models, transformers, and efficient architectures. Research areas include computer vision, large language models with vision, and AI-driven food analytics. Key projects: VISA-DEEP AI chair, Foundation VaViM models, and SmolVLA collaboration with Hugging Face. His recent work examines scalable multimodal models , trajectory prediction , and diffusion-based segmentation , with studies on in-context learning and biased shortcut learning in visual question answering. Articles highlight DeiT variants , fishr for OoD generalization , and STEEX for counterfactual explanations . Scientific awards include IUF Honorary Membership (2009), BMVC 2017 Best Paper, and ICIP 2018 Best Paper. As an advisor, he supervised PhD theses on topics like GAN editing , semantic segmentation , and multimodal retrieval . Current roles involve mentoring the 'Research Band' at MLIA and leading EU-funded initiatives like SCAPE. His work bridges theoretical AI exploration with practical applications in autonomous driving and food technology.

Naoki Yoshinagaمشاهده پروفایل

دانشیار

• Natural Language Processing
• Computational Linguistics
• Multilingual and Multimodal NLP
• +۴ مورد دیگر

Naoki Yoshinaga is a tenured Associate Professor at the Institute of Industrial Science, The University of Tokyo, with extensive experience in natural language processing and computational linguistics. He has held academic positions since 2008 and currently leads research on pragmatic NLP models and multilingual systems. PhD in Computer Science, The University of Tokyo (2005-2008) MSc in Information Science (2000-2002) BSc in Information Science (1996-2000) His research focuses on mechanistic interpretability in NLP models, multilingual/multimodal NLP , and efficient model design using trie structures and conjunctive features. He also investigates knowledge acquisition from social data and evaluation metrics for language generation . Recent publications include work on neuron empirical gradient analysis (ACL-25), multilingual knowledge representation (EACL-24), and compact embedding methods (CoNLL-24). His research has been funded by multiple grants, including the University of Tokyo Excellent Young Researcher program and JSPS fellowships. Committee Special Award, Association for NLP (2023) JSAI SIG Research Award (2022) Best Interactive Award, DEIM Forum (2019, 2016) He developed widely-adopted NLP tools like pecco (fast classification library), RenTAL (LTAG-to-HPSG grammar converter), and J.DepP (Japanese dependency parser). His lab emphasizes strong equivalence in formalism comparisons and pragmatic model design .

Jonathan Shihao Jiمشاهده پروفایل

دانشیار

• Deep Learning
• Computer Vision
• Natural Language Processing
• +۵ مورد دیگر

Jonathan Shihao Ji is an Associate Professor in the School of Computing at the University of Connecticut (UConn), leading the Intelligent Systems Lab. He holds a Ph.D. in Electrical and Computer Engineering from Duke University and previously served as an Associate Professor at Georgia State University and Director of the DoD Center of Excellence (CiARE). His research focuses on deep learning applications in computer vision, NLP, robotics, and high-performance computing, with over 50 publications in top venues like CVPR, NeurIPS, and IEEE journals. He has secured grants from NSF, NIH, DoD, and industry partners including VMware and Nvidia. His work emphasizes efficient algorithms for large-scale data processing, parameter-efficient model fine-tuning (e.g., VB-LoRA), and 3D perception benchmarks for UAVs (UAV3D). Notable contributions include sparse network optimization (Dep-L0), energy-based models (M-EBM), and robust defenses against adversarial attacks (Defense-VAE). He is a Senior Member of IEEE and has developed open-source tools like Parallel Word2Vec and WordRank. Recent projects include accelerating Llama2 models on FPGAs (LlamaF) and improving text-to-image synthesis via contrastive learning. His research spans theoretical advancements and practical applications, with industry collaborations in healthcare, robotics, and embedded systems.

• Computer Vision
• Artificial Intelligence
• Multimodal Learning
• +۷ مورد دیگر

Dr. Hilde Kuehne is a Professor at the University of Tuebingen and a key researcher at the Tuebingen AI Center. She holds affiliations with MIT-IBM Watson AI Lab and Goethe University Frankfurt, with a focus on computer vision, multimodal learning, and explainable AI. Her work bridges vision-language models, audio-visual alignment, and self-supervised methods. Co-organizer of the New Frontiers in Associative Memories workshop @ ICLR 2025 Member of the Scientific Advisory Board of the Carl-Zeiss-Foundation Contributor to CVPR 2025's UTD dataset for unbiased video benchmarks Her research addresses critical challenges in: Explainability for Vision Transformers (LeGrad) Fine-grained audio-visual alignment (CAV-MAE Sync) Zero-shot visual recognition automation (Meta-Prompting) Spatio-temporal grounding without annotations Recent collaborative work spans 15+ publications across CVPR, NeurIPS, ICCV, and ICLR, with emphasis on multimodal foundation models, dataset bias mitigation, and differentiable logic networks. She actively contributes to workshop organization and peer review as evidenced by her involvement in CVPR 2025 and ICLR 2025 program committees.

Michael C. Frankمشاهده پروفایل

استاد

• Developmental Psychology
• Language Acquisition
• Cognitive Development
• +۳ مورد دیگر

Michael C. Frank is the Benjamin Scott Crocker Professor of Human Biology at Stanford University and Director of the Symbolic Systems Program. He leads the Stanford Language and Cognition Lab and has pioneered large-scale collaborative projects including Wordbank (open vocabulary data), MetaLab (developmental meta-analyses), ManyBabies (replication network), childes-db (language transcripts), and Peekbank (eye-tracking repository). His research examines children's language learning and its interaction with social cognition, utilizing computational modeling, large datasets, and open science frameworks. Key interests include: Mechanisms of early language acquisition Pragmatic inference in social contexts Cross-cultural variability in cognitive development Data-driven approaches to developmental science Reproducibility and meta-scientific innovation Recent publications (2022-2025) demonstrate strong emphases on: 1) Novel methods for measuring language environments and cognitive abilities, 2) Computational models of learning and perception, 3) Cross-cultural investigations of social cognition, and 4) Infrastructure for open developmental science. The majority employ multimodal data, meta-analytic techniques, and large-scale collaborations. He teaches courses including Experimental Methods, Developmental Psychology, and interdisciplinary seminars on language, cognition, and computation. His lab maintains active research teams across multiple continents through initiatives like ManyBabies and LEVANTE.

Karen Livescuمشاهده پروفایل

استاد

• Speech Processing
• Language Processing
• Machine Learning
• +۵ مورد دیگر

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.

Benjamin Busamمشاهده پروفایل

پژوهشگر

• 3D Computer Vision
• Multi-Modal Sensor Fusion
• Collaborative Robotics
• +۶ مورد دیگر

Dr. Benjamin Busam is a Senior Research Scientist at the Technical University of Munich , affiliated with the Chair for Computer Science Applications in Medicine under Prof. Nassir Navab. Starting September 2025, he will hold the Professorship for Photogrammetry and Remote Sensing at TUM. His career includes leadership roles at FRAMOS Imaging Systems and Huawei Research in London. Education: Mathematics (TUM), Mathematics and Physics (ParisTech, University of Melbourne), PhD in Mathematics (TUM, 2014) His research focuses on 3D computer vision , multi-modal sensor fusion , and their applications in collaborative robotics and augmented reality . He specializes in projective geometry , 6D pose estimation , and neural radiance fields , with a particular emphasis on photometrically challenging environments. Recent publications highlight advancements in 3D scene understanding , neural rendering , and medical imaging , often leveraging machine learning and vision-language models . His work has been recognized through awards like the EMVA Young Professional Award (2015) and Innovation Pioneer of the Year (2019) , along with multiple Outstanding Reviewer distinctions at leading conferences. Dr. Busam has supervised numerous PhD and MSc students on topics including 6D pose estimation , medical augmented reality , and robotic ultrasound , collaborating with institutions like MIT , École Polytechnique , and University of Padova .

James S. Kimمشاهده پروفایل

استاد

• Early Education
• Education Policy
• Evidence-Based Intervention
• +۷ مورد دیگر

James S. Kim is a Professor of Education at Harvard University's Graduate School of Education, where he conducts policy-relevant research focused on improving literacy outcomes for low-income students and struggling readers. With an Ed.D. from Harvard University (2002), he leads the READS Lab (Research Enhances Adaptations Designed for Scale in Literacy), a research team that partners with school districts to solve literacy challenges through evidence-based interventions. Dr. Kim's research centers on understanding how building students' domain knowledge and reading engagement can foster long-term improvements in reading comprehension. His work emphasizes experimental design and evidence-based interventions, with a particular focus on addressing educational inequality. His research interests include early education, education policy, evidence-based intervention, human development, inequality and education gaps, informal and out-of-school learning, language and literacy development, and teachers and teaching. Kim's most significant contribution is the Model of Reading Engagement (MORE), a spiraled and sustained content literacy intervention co-developed with schoolteachers that has been shown to improve first to third-grade students' reading comprehension in science, English language arts, and math. Notably, research on MORE meets WWC (What Works Clearinghouse) standards without reservation, and long-term follow-up suggests the intervention's impact persists through fourth grade. His publications reveal a consistent focus on content literacy, domain knowledge development, and transfer effects in reading comprehension. Research on MORE meets WWC standards without reservation Long-term effects of MORE persist through fourth grade Commitment to Open Science principles (open data, open materials, preregistration) As a servant leader, Kim builds long-term partnerships with school districts to implement literacy interventions at scale. His READS Lab promotes open science practices while developing practical solutions to literacy challenges. His research on summer reading interventions, parental text messaging, and classroom-based content literacy approaches demonstrates his commitment to translating research into practice. Kim's current work includes scaling the MORE intervention to improve reading comprehension for high-needs students in moderate to high poverty schools through a Department of Education-funded project (2024-2028).

• Computer Vision
• Generative Models
• Image-Text Joint Modeling
• +۵ مورد دیگر

Svetlana Lazebnik is a Full Professor and Willett Faculty Scholar in the Department of Computer Science at the University of Illinois at Urbana-Champaign (UIUC), part of the Grainger College of Engineering. She holds a Ph.D. from UIUC (2006) and previously served as an Assistant Professor at the University of North Carolina at Chapel Hill (2007–2011). Her research focuses on computer vision, including generative models for virtual try-on, image stylization, scene understanding, and joint modeling of images and language. She has advised numerous Ph.D. students and postdocs, many of whom now hold prominent academic and industry roles. Education: Ph.D. in Computer Science, UIUC (2006); supervised by Jean Ponce. Research Interests: Her work spans generative adversarial networks (GANs), diffusion models, virtual try-on systems (e.g., Dressing-in-Order, Street Try-On), exemplar-based stylization, and large-scale photo analysis. She has pioneered spatial pyramid matching and contributed to binary code learning for image retrieval. Key Awards: NSF CAREER Award (2008), Microsoft Research Faculty Fellow (2009), Sloan Research Fellow (2013), IEEE Fellow (2021), and the Longuet-Higgins Prize (2016) for her CVPR 2006 paper. Teaching: Recent courses include CS 444 (Deep Learning for Computer Vision), CS 543 (Computer Vision), and a Ph.D. Job Search Seminar. She has also taught at UNC Chapel Hill. Grants & Funding: Supported by NSF, Amazon, AWS, Microsoft, Sloan Foundation, Google, ARO, and Adobe. Notable grants include CCF 2348624 and IIS 1718221. Labs/Groups: Leader in the Illinois CS Vision Group, contributing to collaborative projects on embodied AI, multi-agent systems, and visual-semantic reasoning.

Lifeng Zhouمشاهده پروفایل

استادیار

• Robotics
• Algorithms
• Combinatorial Optimization
• +۴ مورد دیگر

Lifeng Zhou is an Assistant Professor in the Department of Electrical and Computer Engineering at Drexel University, where he leads the Zhou Lab focused on advancing robustness and reliability in multi-robot systems through integration of foundation models. His research addresses real-world challenges in environmental monitoring, disaster response, and urban mobility. Education PhD, Electrical and Computer Engineering, Virginia Tech, 2020 MS, Control Science and Engineering, Shanghai Jiao Tong University, 2016 BS, Automation, Huazhong University of Science and Technology, 2013 Research Focus Dr. Zhou's work integrates robotics, algorithms, game theory and machine learning to develop secure and scalable autonomous systems. Primary research thrusts include: Resilient multi-robot coordination in adversarial environments Large language model integration for robotic decision-making Game-theoretic resource allocation strategies Risk-aware planning for autonomous vehicles Publication Trends Recent work (2024-2025) demonstrates strong focus on large language model applications in multi-robot systems, with 12/15 articles exploring LLM integration for flocking, scene segmentation, and decision-making. Additional emphasis includes adversarial robustness in target tracking (5 articles) and autonomous driving applications (4 articles). Awards and Recognition Best Paper Award, WACV 2025 LLVM-AD Workshop Professional Service Associate Editor, ICRA Conference Editorial Board Laboratory Focus The Zhou Lab develops foundational algorithms for secure and scalable multi-robot systems, with current projects spanning environmental monitoring drones, disaster response coordination, and autonomous vehicle perception systems.

Manolis Savvaمشاهده پروفایل

دانشیار

• Computer Graphics
• Computer Vision
• 3D Scene Analysis
• +۴ مورد دیگر

Manolis Savva is an Associate Professor in the School of Computing Science at Simon Fraser University and holds the Canada Research Chair in Computer Graphics. He specializes in 3D scene analysis, generative methods for 3D content, and computer graphics for AI. His research bridges computer graphics, vision, and robotics. Education: Ph.D. (Computer Science, Stanford University, 2016), MS (Computer Science, Stanford, 2012), B.A. (Physics & Computer Science, Cornell, 2009). Research Interests: Human-centric 3D scene analysis, generative 3D content creation, AI-driven rendering, and applications in robotics. Key projects include Habitat (Embodied AI platform), ScanNet , and ShapeNet datasets. Recent Articles: Focus on articulated object modeling, 3D scene synthesis, and AI-driven visualization. Notable work includes SceneMotifCoder (generating object arrangements) and R3DS (panoramic scene understanding). Awards: CHCCS Early Career Award (2022), ICLR 2023 Outstanding Paper Award, ICCV 2019 Best Paper Nomination, and SGP 2020 Dataset Award (ScanNet). Lab/Teams: Leads research groups in 3DLG (3D Learning and Graphics) and GrUVi (Graphics and Vision). Collaborates on projects like AI Habitat and HomeRobot .