Find Academic Supervisors

Wei XuView profile

Associate Professor

• Large Language Models
• Natural Language Processing
• Machine Learning
• +6 more

Wei Xu is an Associate Professor at Georgia Institute of Technology's College of Computing and School of Interactive Computing, with affiliations to the Machine Learning Center. Their research bridges machine learning, natural language processing, and social media with focus areas in large language models, cultural bias mitigation, multilingual capabilities, and human-AI collaboration in text evaluation. NSF CAREER and Google Academic Research Award recipient Director of NLP X Lab PhD from New York University, BSMS from Tsinghua University Research interests span: Multilingual Multicultural LLMs addressing representational gaps and cultural adaptation in language models (NAACL 2025, ACL 2024); Robustness and Reasoning through dynamic AGI evaluations (ACL 2024, EMNLP 2024); Interdisciplinary NLP applications in security, healthcare, and law (EMNLP 2024, ACL 2024). Recent publications focus on multilingual alignment (NAACL 2025), privacy risk estimation (arXiv 2025), cultural bias analysis (ACL 2024), and medical text simplification (EMNLP 2024). Key themes include bias mitigation, multimodal processing, and practical LLM evaluation. Scientific Awards : NSF CAREER, Google/Sony/Criteo research awards, ACL'24 Best Social Impact Award, COLING'18 Best Paper Advising 15 PhD/MS/BSMS students including Yao Dou (human-centered LLM evaluation), Tarek Naous (multilingual LLMs), and alumni like Chao Jiang (Apple AI/ML) and Yang Chen (NVIDIA research scientist). Teaches graduate courses on NLP and LLMs.

Wojciech MatusikView profile

Professor

• Computer Graphics
• Computational Design and Fabrication
• Computer Vision
• +7 more

Wojciech Matusik is a Professor of Electrical Engineering and Computer Science at MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL). He leads the Computational Design and Fabrication Group and is a member of the Computer Graphics Group. His research spans computer graphics, robotics, and AI-driven manufacturing, with a focus on computational design, tactile sensing, and material science. Matusik holds a PhD in Computer Science from MIT (2003), an MS from MIT (2001), and a BS from UC Berkeley (1997). His work includes groundbreaking projects like differentiable cloth simulation (DiffCloth), AI-enhanced molecular design, and tactile sensing gloves. He has received prestigious awards such as the MIT TR35 (2004), DARPA Young Faculty Award (2012), and Ruth and Joel Spira Teaching Award (2014). Matusik teaches courses on computer graphics, machine learning, and computational fabrication at MIT. Key research themes include: Robotics: Robotic assembly, tactile interaction, and soft robotics Graphics: 3D holography, procedural material generation Manufacturing: Additive fabrication, topology optimization His recent articles explore AI-driven molecular synthesis, holographic displays, and tactile-enabled VR systems. Matusik collaborates on open-source tools like the WiReSens tactile platform and Simit language for sparse systems.

• Computer Vision
• Machine Learning
• Data-Driven Learning
• +5 more

Alexei A. Efros is the Howard Friesen Professor in the EECS Department at the University of California, Berkeley, and a core member of the Berkeley Artificial Intelligence Research (BAIR) Lab. Previously, he spent a decade at Carnegie Mellon University’s Robotics Institute. His research focuses on data-driven computer vision, self-supervised learning, computational photography, and generative models. He has pioneered advancements in visual representation learning, including seminal work on neural radiance fields and generative adversarial networks. Education Background: Efros holds a PhD in Computer Science from MIT, though specific details of his academic journey are not explicitly provided in the text. His career includes postdoctoral research at the University of Oxford with Andrew Zisserman and collaborative work with Team WILLOW at INRIA Paris. Research Interests: Efros explores how vast uncurated visual data can be leveraged for understanding and synthesizing the visual world. Key areas include self-supervised learning, generative models, and applications in robotics and art. His lab has contributed influential techniques such as Style Transfer, GAN-based image synthesis, and neural scene representation learning. Recent work emphasizes real-time adaptation (Test-Time Training), 3D perception models, and ethical AI implications of generative systems. Publications: Over 150+ publications span topics like Generative Adversarial Networks (GANs), unsupervised learning, and visual-linguistic models. Notable works include Unpaired Image-to-Image Translation (CUT/GAU), Style Transfer , and Swapping Autoencoder . His research has significant industry impact, with techniques adopted in Adobe’s software and generative AI applications. Grants & Collaborations: Efros has secured major funding from NSF, DARPA, and industry partnerships (e.g., Adobe, NVIDIA). He co-leads projects on scalable vision models, ethical AI, and real-world perception systems. Current collaborations include work with MIT, NYU, and INRIA Paris. Labs & Teams: Leads the BAIR Vision Group at Berkeley, fostering interdisciplinary research between computer vision, graphics, and robotics. The group emphasizes Slow Science principles, prioritizing deep exploration over rapid publication.

Larry PileggiView profile

Professor

• Secure Integrated Circuit Hardware
• Integrated Circuits Design Methodologies
• Power Systems Simulation and Optimization
• +3 more

Larry Pileggi is the Coraluppi Head and Tanoto Professor of Electrical and Computer Engineering at Carnegie Mellon University (CMU). He is also the Department Head of CMU’s ECE Department and has held prior roles at Westinghouse Research and Development and the University of Texas at Austin. His work bridges academic research and industry innovation, with co-founding ventures like Fabbrix Inc., Extreme DA, and Pearl Street Technologies. Dr. Pileggi earned his Ph.D. in Electrical and Computer Engineering from CMU (1989), following an M.S. (1984) and B.S. (1983) in Electrical Engineering from the University of Pittsburgh. His research focuses on three core areas: secure integrated circuit hardware (mitigating supply-chain threats), integrated circuits design methodologies (supporting sub-20nm CMOS and heterogeneous technologies), and power systems simulation (developing robust grid analysis tools like SUGAR). He has been recognized with numerous awards, including the prestigious 2023 Phil Kaufman Award for contributions to electronic system design, and is an IEEE Fellow. His academic leadership includes fostering maker initiatives and interdisciplinary research programs to address future challenges in energy and computing systems. Pileggi’s advising record includes over 47 Ph.D. students, many of whom collaborate with him in industry ventures. Current and past grants support his work on resilient power grids and novel memory technologies, reflecting a commitment to both theoretical and applied research. His lab, the Pileggi Lab, develops cutting-edge solutions for energy and integrated systems, emphasizing scalable simulation, secure hardware design, and next-generation memory architectures. Collaborations span institutions like ETH Zurich and industry partners in EDA and semiconductor sectors.

Manolis SavvaView profile

Associate Professor

• Computer Graphics
• Computer Vision
• Robotics
• +4 more

Roles & Affiliations: Manolis Savva is an Associate Professor at Simon Fraser University's School of Computing Science and holds a Canada Research Chair in Computer Graphics. He leads research in 3D scene understanding, with applications in graphics, vision, and robotics. Previously, he was a researcher at Facebook AI and Princeton University. Education: Ph.D. in Computer Science (2016), Stanford University, advised by Pat Hanrahan B.A. in Physics and Computer Science (2009), Cornell University Research Interests: His work focuses on analyzing, organizing, and generating 3D content, particularly for holistic scene understanding. Key areas include articulated objects, embodied AI, and datasets like ScanNet , Matterport3D , and Habitat . His methods drive applications in robotics, autonomous agents, and virtual environments. Publications Trends: Recent work emphasizes generative models (e.g., SINGAPO for articulated object parts), embodied AI benchmarks (Habitat), and multimodal scene analysis. His papers often address challenges in scalability, realism, and cross-modal fusion for 3D environments. Awards: CHCCS Early Career Researcher Award (2022) ICLR 2023 Outstanding Paper Award ICCV 2019 Best Paper Nomination (Habitat) Advising & Grants: Supervised over 15 graduate students, many advancing to top PhD programs and tech companies. Active in grants for embodied AI, scene understanding, and robotics. Labs/Teams: Leads the 3DLG (3D Learning Group) and GrUVi (Graphics and Vision) groups at SFU. Collaborates extensively with industry (e.g., Meta, NVIDIA) on AI-driven 3D research.

Giovanni De MicheliView profile

Professor

• Integrated Circuit Design
• Hardware Synthesis
• Logic Optimization
• +5 more

Giovanni De Micheli is a Full Professor at École Polytechnique Fédérale de Lausanne (EPFL), holding appointments in both the School of Computer and Communication Sciences (IC) and the School of Engineering (STI). He is affiliated with the Integrated Systems Laboratory (LSI) and serves as Scientific Director of EcoCloud. His primary office is located in INF 341 building at EPFL's Lausanne campus where he conducts research and teaching activities. Professor De Micheli's research spans multiple domains in electronic design automation and integrated circuit design. His primary interests include: Integrated Circuit Design and Hardware Synthesis Logic Optimization and Boolean Methods Nanoelectronics and 2D Electronics Quantum Computing and Emerging Technologies Low-Power Electronics and Energy-Efficient Systems Electronic Design Automation for Novel Computing Paradigms His recent publications demonstrate a strong focus on advancing logic synthesis techniques for both conventional and emerging technologies. Professor De Micheli's work bridges theoretical foundations with practical applications, particularly in adapting traditional EDA methods to novel computing paradigms like quantum computing and superconducting electronics. His research group has made significant contributions to the development of efficient algorithms for circuit optimization across multiple technology domains, with particular emphasis on area, power, and delay optimization while maintaining functional correctness. Professor De Micheli has supervised numerous doctoral students throughout his career at EPFL, mentoring over 40 PhD candidates whose research spans various aspects of electronic design automation, circuit design, and emerging technologies. His students have gone on to make significant contributions in both academia and industry. As Scientific Director of EcoCloud, Professor De Micheli leads research initiatives focused on energy-efficient computing systems and sustainable cloud infrastructure. His laboratory, the Integrated Systems Laboratory, serves as a hub for interdisciplinary research connecting computer science, electrical engineering, and materials science, with particular focus on next-generation computing technologies.

Robert MacCurdyView profile

Assistant Professor

• Automated robot design
• Additive manufacturing
• Soft robotics
• +6 more

Robert MacCurdy is an Assistant Professor at the Department of Mechanical Engineering, University of Colorado Boulder . He leads the Matter Assembly Computation Lab (MACLab) focused on automating robot design and fabrication. His research bridges computational design and advanced manufacturing to create "robots that walk out of the printer." The lab develops tools like OpenVCAD , an open-source volumetric multi-material geometry compiler.

• Computer Vision
• Machine Learning
• Robotics
• +4 more

Andreas Geiger is a Professor and Head of the Department of Computer Science at the University of Tübingen, Germany. He leads the Autonomous Vision Group (AVG) within CyberValley and is a core faculty member of the Tübingen AI Center. His roles also include PI in the ML in Science Excellence Cluster and the CRC Robust Vision, as well as ELLIS Fellow and coordinator of the ELLIS PhD program. He specializes in machine learning models for computer vision, robotics, and autonomous systems, with applications in self-driving cars, VR/AR, and scientific document analysis. Educational background: While not explicitly detailed, his positions imply a Ph.D. in Computer Science or related field. His work spans interdisciplinary collaborations with institutions like ETH Zürich, Microsoft, and the University of Bonn. Research focuses on 3D scene understanding, Gaussian splatting, generative models, and reliable autonomous systems. Notable contributions include the KITTI dataset and foundational work in neural radiance fields. Awards include the Sage 10-Year Impact Award (2024), ERC Starting Grant (2019), and IEEE PAMI Young Researcher Award (2018). Key projects include the Scholar Inbox paper recommender platform, ReSim (reliable world simulation), and advancements in 3D scene generation (e.g., UrbanCAD, PrITTI). His lab maintains a strong focus on open-source tools and datasets, such as the CARLA Route Generator. Grants and funding include support from Vector Stiftung (MINT innovation program) and EU initiatives like the ML in Science Cluster. His team collaborates internationally, with recent work presented at CVPR, SIGGRAPH, and NeurIPS.

• Computer Engineering
• Circuit Design
• Fault Tolerance
• +6 more

Matteo Sonza Reorda is a Full Professor at the Polytechnic University of Turin, affiliated with the Department of Automatic Control and Computer Science (DAUIN). He holds roles such as Partnership Agreement Coordinator with SMAT and has served as Vice-Rector for Research (2021–2024). His research focuses on fault tolerance, hardware reliability, and AI acceleration, with contributions to GPU/CNN reliability, self-test libraries, and defect-oriented testing. Education: M.S. in Electronics Engineering, Politecnico di Torino (1986) Ph.D. in Computer Engineering, Politecnico di Torino (1990) Research Interests: His work spans automatic test equipment, circuit reliability, GPU fault tolerance, neural network hardening, and embedded systems safety . He leads the Electronic CAD & Reliability Group and collaborates with institutions like the ISI Foundation and the National Research Council (CNR). Publications & Impact: With over 180 publications, his recent work emphasizes AI accelerator reliability, fault injection techniques, and safety-critical systems. Key trends include GPU resilience for CNNs, self-test library optimization, and radiation fault modeling in neural networks. Awards & Recognition: IEEE Fellow (2016–) Multiple best-paper awards at IEEE conferences (DDECS, DATE, etc.) Grants & Collaborations: Coordinator of EU projects (RESCUE, TUTORIAL) Partnerships with EDF, Marelli Europe, and SMAT Lead on National HPC/Quantum Computing Spoke 1 (2022–2025) Labs/Teams: Leads the CAD Group and collaborates with the French-Italian LIA LAFISI lab on hardware-software integration.

Adrian LewView profile

Professor

• Computational Solid Mechanics
• Numerical Algorithms
• Hydraulic Fracturing Simulation
• +3 more

Adrian Lew is a Professor of Mechanical Engineering at Stanford University, specializing in computational solid mechanics and numerical algorithms. His research focuses on hydraulic fracturing simulation, embedded boundary methods, and material model design. He holds a PhD in Mechanical Engineering from Caltech (2003). His work bridges advanced numerical techniques with real-world applications in geophysics, material science, and structural engineering. Education: PhD, Mechanical Engineering, California Institute of Technology, 2003 Research Interests: Lew's group develops algorithms for time-integration embedded boundary methods and hydraulic fracturing simulations. Key areas include curvilinear crack propagation, universal meshing for complex geometries, and high-fidelity fracture mechanics. His work on variational integrators and discontinuous Galerkin methods has advanced computational efficiency in nonlinear elasticity and thermodynamics. Publications: Recent articles emphasize mesh optimization (DVRlib), fracture path instabilities, and magma chamber dynamics. His methodologies address challenges in 3D crack modeling, fluid-structure interaction, and high-order approximations in domains with singularities. Advising & Grants: Lew's research is supported by projects in computational geophysics and material science. Though no advisees are listed, his work involves collaborative teams focused on algorithmic innovation and high-performance computing.

Morteza FayaziView profile

Assistant Professor

• Electronic Design Automation (EDA)
• Machine Learning (ML) in Analog Circuit Design
• System-on-Chip (SoC) Design Automation
• +3 more

Morteza Fayazi is an Assistant Professor in the Department of Electrical and Computer Engineering at the University of Utah, with an adjunct position in the Kahlert School of Computing. His research focuses on Electronic Design Automation (EDA), applying machine learning to automate analog and mixed-signal circuit design, and developing high-performance computing systems. He holds a B.Sc. from Sharif University of Technology, and M.S.E./Ph.D. degrees from the University of Michigan. His research interests include AI-driven EDA, RF/circuit automation, and energy-efficient processors. Key achievements include the MEDAL lab’s work on terahertz radars, systolic-array processors (e.g., DAP and Versa), and open-source frameworks like FASCINET and Tablext. He has received awards such as the 2024 College of Engineering Dean’s ETR Fund and the 2017 Outstanding Undergraduate Thesis Award. Teaching responsibilities include multiple iterations of the Digital System Design course (ECE/CS 3700). His work spans over 15 peer-reviewed articles in IEEE Transactions, ACM, and top conferences like ICCAD and VLSI-SOC, emphasizing automation, efficiency, and AI integration in hardware design.

• Photonics
• Neuromorphic Engineering
• Optical Communications
• +5 more

Paul Prucnal is a Professor of Electrical and Computer Engineering at Princeton University, affiliated with the Princeton Materials Institute (PMI). He leads the Lightwave Communications Research Lab, focusing on ultrafast optical techniques for communication networks and neuromorphic photonics. His research spans optical security, CDMA networks, nonlinear signal processing, and photonic neurons. Education: Ph.D., Columbia University (1979) M.Phil., Columbia University (1978) M.S., Electrical Engineering, Columbia University (1976) A.B., Bowdoin College, summa cum laude (1974) Research Interests: Optical Network Security (eavesdropping/jamming countermeasures) Optical CDMA for broadband networks Silicon photonic neuromorphic computing RF interference cancellation in wireless systems Photonic spiking neurons mimicking biological organisms Awards: National Academy of Inventors Fellow (2017) 10+ teaching awards including Princeton's President's Award (2015) OSA/IEEE Fellowships (1992, 1997) Labs/Teams: Leads the Lightwave Communications Lab, collaborating with government/industry partners. Lab alumni like Prof. Bhavin Shastri have achieved international recognition. Grants/Publications: Over 350 journal papers, 22 U.S. patents. Authored/co-authored Neuromorphic Photonics (2017) and edited Optical Code Division Multiple Access (2019). Current projects include photonic tensor processors and real-time RF signal processing.

Levent Burak KaraView profile

Professor

• Machine Learning
• Artificial Intelligence
• Computational Design
• +12 more

Levent Burak Kara is a Professor in the Department of Mechanical Engineering at Carnegie Mellon University (CMU), with a courtesy appointment in the Robotics Institute. He is a leading researcher in AI-driven computational design, additive manufacturing, and intelligent engineering systems, leading the Visual Design and Engineering Lab (VDEL) at CMU. Education: B.S., Mechanical Engineering, Middle East Technical University (1998) M.S., Mechanical Engineering, Carnegie Mellon University (2000) Ph.D., Mechanical Engineering, Carnegie Mellon University (2005) His research focuses on integrating machine learning, optimization, and geometric modeling to revolutionize engineering design and manufacturing. Key areas include topology optimization, CAD intelligence, digital twins, generative design, bioengineering, and electronic design automation. His work enables automation of traditionally labor-intensive design processes using deep learning and reinforcement learning. His recent publications reveal a strong trend toward physics-informed surrogate modeling, real-time simulation, manufacturability prediction, and AI-driven automation in mechanical, biomedical, and electronic systems. These works frequently appear in top journals such as Journal of Mechanical Design and Journal of Applied Mechanics , and at premier conferences like NeurIPS and DAC. Scientific Awards: National Science Foundation CAREER Award ASME Design Automation Society Young Investigator Award Google AI for Social Good Impact Scholar Kara advises several Ph.D. students and has secured significant funding from federal agencies such as the NSF and the U.S. Army Research Laboratory, as well as collaborations with industrial leaders including Cadence Design Systems and NVIDIA. His research is also supported by CMU’s NextManufacturing Center and the Critical Technology Initiative. He is actively involved in developing intelligent design systems that leverage AI to automate product design, optimize manufacturing processes, and improve medical diagnostics, particularly in oral cancer screening and organ preservation. His lab, VDEL, is a hub for innovation in AI-enabled engineering.

Guglielmo ScovazziView profile

Professor

• Finite element methods
• Computational fluid mechanics
• Computational solid mechanics
• +9 more

Guglielmo Scovazzi is a Professor at Duke University with appointments across multiple departments including the Department of Civil and Environmental Engineering, the Thomas Lord Department of Mechanical Engineering and Materials Science, and as Professor of Mathematics. His interdisciplinary research bridges computational mechanics, scientific computing, and engineering applications. Dr. Scovazzi earned his B.S/M.S. in aerospace engineering (summa cum laude) from Politecnico di Torino (Italy), followed by an M.S. and Ph.D. in mechanical engineering from Stanford University. Prior to joining Duke, he was a Senior Member of the Technical Staff at Sandia National Laboratories' Computer Science Research Institute. His research focuses on developing advanced numerical methods for computational mechanics, particularly finite element methods for fluid and solid mechanics. Key areas include multiphase porous media flows, computational methods for materials under extreme conditions, turbulent flow computations, and instability phenomena. His work emphasizes creating accurate computational approaches that reduce design/analysis costs for complex engineering problems involving fluid-structure interactions and transient phenomena in complex geometries. Dr. Scovazzi's most significant recent contribution is the development of the Shifted Boundary Method, an innovative computational framework that enables efficient simulations on complex geometries without requiring boundary-fitted meshes. This method has found applications in geomechanics, energy systems, and resilient infrastructure design. Kavli Fellow, National Academy of Sciences & Kavli Foundation (2018) Presidential Early Career Award for Scientists and Engineers (PECASE), White House (2017) Early Career Award, U.S. Department of Energy, Advanced Scientific Computing Research Program (2014) Dr. Scovazzi teaches multiple courses in computational mechanics including Nonlinear Finite Element Analysis and Introduction to the Finite Element Method. His research has been supported by substantial federal funding, and he actively collaborates across disciplines to address challenging problems in energy, environment, and infrastructure resilience through advanced computational methods.

Faez AhmedView profile

Associate Professor

• Generative design
• Machine learning and optimization for engineering design
• Human-AI collaboration
• +4 more

Faez Ahmed is an Associate Professor at the Department of Mechanical Engineering, Massachusetts Institute of Technology (MIT), where he serves as the Doherty Chair in Ocean Utilization. He leads the Design Computation and Digital Engineering (DeCoDE) Lab, focusing on integrating machine learning and optimization with engineering design to enhance human-AI collaboration and accelerate design processes. Ph.D., Mechanical Engineering, University of Maryland College Park (2019) B.Tech.-M.Tech., Mechanical Engineering, Indian Institute of Technology Kanpur (2012) His research interests include generative design methodologies, AI-driven optimization techniques, and the development of algorithms that facilitate collaboration between human designers and artificial intelligence systems. This interdisciplinary work spans applications in automotive design , ship hull synthesis , and wind turbine optimization , with a strong emphasis on creating open-source tools and datasets for the engineering community. Recent publications demonstrate his lab's leadership in fields such as 3D CAD generation , multimodal design datasets , and constraint-aware generative models . These works often address challenges in design space exploration , performance prediction , and data-driven design frameworks . Scientific Awards NSF CAREER Award (2025) ASME Young Investigator Award (2024) Google Research Scholar Award (2024) 3M Non-Tenured Faculty Award (2022) University of Maryland Alumni Research Award (2022) Faez Ahmed's lab has trained numerous Ph.D. candidates and postdoctoral researchers, fostering a collaborative research environment that bridges mechanical engineering , artificial intelligence , and computational methods . The DeCoDE Lab actively engages with industry partners and academic institutions, contributing to large-scale datasets and benchmarks that power the next generation of engineering design research.