جستجوی استاد راهنمای دانشگاهی

Greg Durrett is an Associate Professor in the Department of Computer Science at University of Texas at Austin, leading the TAUR Lab (Text Analysis, Understanding, and Reasoning ). His research focuses on advancing Large Language Models (LLMs) for knowledge-intensive tasks in medical information processing scientific discovery legal reasoning . He received his B.S. in Computer Science and Mathematics from MIT (2010) and Ph.D. in Computer Science from UC Berkeley (2016). His work develops techniques to train LLMs with new capabilities augment models for reliability assess model outputs improve reasoning frameworks . His 15 most recent publications (2021-2025) span knowledge propagation in LLMs chain-of-thought reasoning code generation benchmarks multi-modal reasoning fact verification discourse analysis . Scientific honors include NSF CAREER Award (2024) NSF grants (2018, 2024) Bloomberg Data Science Grant (2017) Facebook Fellowship (2014) Best Paper Finalist (EMNLP 2013) . Teaching: CS388: Natural Language Processing (graduate) CS371N: NLP (undergraduate) High school NLP module .

• Natural Language Processing
• Machine Learning
• Artificial Intelligence
• +۷ مورد دیگر

Tengyu Ma is an Assistant Professor of Computer Science at Stanford University. His research focuses on machine learning, deep learning, optimization, and theoretical computer science. He is particularly known for work on neural networks, reinforcement learning, and algorithmic guarantees in AI systems. His email is tengyuma@stanford.edu . Ma's research interests span foundational aspects of machine learning, including generalization theory, optimization algorithms, and the theoretical underpinnings of deep learning. He has contributed to areas such as self-play theorem provers, learning rate schedules, and robustness in low-light vision tasks. His work often bridges theoretical insights with practical algorithm design. His recent publications emphasize advancements in large language models (LLMs), theorem proving via self-play, and understanding training dynamics in deep networks. Despite prolific output, no specific scientific awards are explicitly mentioned in the provided texts. Ongoing work includes exploring in-context learning mechanisms, formal verification of AI systems, and efficient pretraining techniques. His research has implications for both theoretical understanding and real-world applications of AI.

• Machine Learning
• Deep Learning
• Reinforcement Learning
• +۳ مورد دیگر

Nadia Polikarpova is an Associate Professor in the Department of Computer Science and Engineering at the University of California, San Diego . She earned her PhD from ETH Zurich in 2014 under Bertrand Meyer , followed by postdoctoral research at MIT CSAIL with Armando Solar-Lezama . Her academic contributions have been recognized with prestigious awards including the 2020 Sloan Fellowship , 2020 Intel Rising Stars Award , and 2020 NSF CAREER Award . Polikarpova's research focuses on program synthesis , program verification , and type systems . She leads the Programming Systems group at UCSD and contributes to the IFIP Working Group 2.8 on Functional Programming since 2022. Her work spans foundational research and practical tools, including projects like Synquid , SuSLik , and Laurel that combine formal methods with machine learning for code generation. Her recent publications in venues like OOPSLA , NeurIPS , and ICFP reveal trends in AI-assisted programming , live programming environments , and formal verification . She has advised numerous PhD and Master’s students including Shraddha Barke , Zheng Guo , and Tristan Knoth , many of whom have moved to prominent academic and industry positions. Notable artifacts from her lab include tools like ColDeco for spreadsheet inspection and Superfusion for eliminating intermediate data structures. 2020 : Sloan Fellow 2020 : Intel Rising Stars Award 2020 : NSF CAREER Award 2021 : Distinguished Paper at POPL 2023 : Distinguished Artifact at PLDI 2023 : Distinguished Paper at OOPSLA Polikarpova actively contributes to academic service, serving on program committees for PLDI , POPL , and OOPSLA , and co-chairing the OOPSLA Review Committee in 2023. She has delivered keynotes at APLAS'20 and PLDI'24 , emphasizing the integration of large language models with formal methods.

• Program Synthesis
• Program Verification
• Type Systems
• +۵ مورد دیگر

Assia Mahboubi is a tenured researcher ( directrice de recherche ) at INRIA in the Gallinette team, Nantes, France, and an endowed professor in the Algebra and Number Theory section of the Vrije Universiteit Amsterdam, Netherlands. Her work bridges theoretical computer science and formal mathematics, with significant contributions to proof assistants and formal verification. Her research focuses on the foundations and formalization of mathematics in type theory, particularly on the automated verification of mathematical proofs. She explores the interplay between computer algebra and formal proofs, and is a key contributor to the Rocq prover (formerly Coq) and the Mathematical Components libraries. Her work often examines how familiar mathematical objects can be optimally represented for computer-aided proof checking. Recent publications show a strong trend toward categorical reasoning, diagram chasing, and continuity properties in constructive type theory, with increasing focus on practical applications of formal methods in computational mathematics. Her work demonstrates the maturation of formal verification techniques from theoretical foundations to practical tools for mathematical research. ERC Consolidator grant for the FRESCO (Fast and Reliable Symbolic Computation) project Mahboubi actively supervises doctoral students including Vojtěch Štěpančík, Tomás Vallejos Parada, and Alain Chavarri Villarello. She has received significant research funding through her ERC Consolidator grant for the FRESCO project, which aims to develop fast and reliable symbolic computation techniques. She is deeply involved in the international research community, serving on program committees for major conferences including POPL, CPP, and ICFP. She leads research in the Gallinette team at INRIA, which focuses on the intersection of proof assistants, programming languages, and formal mathematics. Her work has helped establish formal verification as a practical tool for mathematical research, moving beyond theoretical foundations to real applications in computational mathematics.

• Type Theory
• Formal Verification
• Computer-Aided Mathematics
• +۳ مورد دیگر

Sanjit A. Seshia is the Cadence Founders Chair Professor in the Department of Electrical Engineering and Computer Sciences at the University of California, Berkeley . He is affiliated with the Group in Logic and the Methodology of Science and participates in centers like the Industrial Cyber-Physical Systems Center , Berkeley AI Research , and the Simons Institute for the Theory of Computing . Research interests include formal methods for automated verification and synthesis of dependable systems, with applications to cyber-physical systems , AI-based autonomy , and computer security . His work spans SMT solving, model counting, syntax-guided synthesis, and algorithmic improvisation, with tools like UCLID5 , VerifAI , and Scenic for verifying autonomous systems and educational platforms like CPSGrader . Students and collaborators include notable researchers such as Dorsa Sadigh (Stanford), Daniel Fremont (UC Santa Cruz), and Hazem Torfah (Chalmers). He has co-founded startups like Decyphir and 20ⁿ Labs based on his research.

• Formal Methods
• Cyber-Physical Systems
• Computer Security
• +۱۰ مورد دیگر

Sean Welleck is an Assistant Professor at Carnegie Mellon University's School of Computer Science, Language Technologies Institute, leading the L3 Lab. His research focuses on bridging informal and formal reasoning with AI, spanning machine learning for mathematics and code, inference algorithms, and AI agents. PhD in Computer Science from New York University (advised by Kyunghyun Cho) Postdoctoral work at University of Washington (advised by Yejin Choi) His work explores AI-driven formal methods for mathematics and code generation, test-time compute scaling, and algorithms enabling AI improvement over time. Recent publications analyze reasoning evaluation, premise selection, and automated proof optimization in systems like Lean. Key article trends include neural theorem proving, code generation, and inference-time compute optimization. Awards: NVIDIA AI Labs Pioneering Research Awards (2017, 2018), NAACL 2025 Best Paper. Current advisees include PhD students Pranjal Aggarwal, Weihua Du (co-advised with Yiming Yang), Andre He (co-advised with Daniel Fried), and Seungone Kim (co-advised with Graham Neubig). He co-organizes workshops like Autoformalization for the Working Mathematician (ICERM 2025) and VerifAI: AI Verification in the Wild (ICLR 2025), and teaches Advanced NLP at CMU.

• Machine Learning for Mathematics
• Machine Learning for Code
• AI Reasoning and Agents
• +۶ مورد دیگر

Josef Urban is a leading researcher at the Czech Institute of Informatics, Robotics and Cybernetics (CIIRC) , Czech Technical University in Prague, heading the ERC Consolidator project AI4REASON . Previously, he held positions as a postdoc at Radboud University Nijmegen and assistant professor at Charles University in Prague, where he co-founded the Prague Automated Reasoning Group. Education Ph.D. in Computer Science (2004), Charles University, Prague M.S. in Mathematics (1998), Charles University, Prague B.S. in Economics (1995), Charles University, Prague Research Interests Urban specializes in automated reasoning over large formalized knowledge bases, combining deductive theorem proving and inductive machine learning . His work aims to realize "strong AI" through formalized mathematics, particularly using systems like Mizar and the AI/TP Challenges . He advocates for computer-verifiable mathematics as a foundation for AI progress. Article Trends Urban's publications focus on integrating machine learning with automated theorem proving in systems like ENIGMA and BliStr . Key trends include semantic guidance for ATPs, premise selection in formal libraries, and automated proof compression via concept invention. Scientific Contributions Head of ERC Consolidator project AI4REASON Marie-Curie Fellow at University of Miami Co-founder of Prague Automated Reasoning Group Editor for Formalized Mathematics Advising and Grants Urban has advised numerous PhD and MSc students including Daniel Kuehlwein, Krystof Hoder, and Yutaka Nagashima. He has secured grants like the ERC Consolidator Grant and Marie-Curie Fellowship . Labs and Collaborations Urban leads the AI4REASON team at CIIRC and collaborates with the Foundations Group at Radboud University. He contributes to projects like Mizar TWiki and XML-based API for Mizar , aiming to create a semantic AI ecosystem for formal knowledge.

• Automated Reasoning
• Machine Learning
• Theorem Proving
• +۵ مورد دیگر

Daniel Balasubramanian is an Adjunct Associate Professor of Computer Science and Research Scientist at Vanderbilt University's School of Engineering. His research focuses on cybersecurity, software verification, and cyber-physical systems, with expertise in symbolic execution, code analysis, and formal methods. He contributes to advancing secure systems through frameworks like RAMPART for adversarial defense and Syntheto for formal verification. His work intersects edge computing, hardware security, and autonomous systems resilience. Research Interests: Cybersecurity (including ethical hacking, network defense), formal methods (verification, theorem proving), edge computing (tinyML, cloud integration), and cyber-physical systems (emulation, testbeds). His recent work emphasizes assurance provenance in software documentation and adversarially robust autonomous systems. Publications since 2019 highlight contributions to cybersecurity testbeds, reinforcement learning for penetration resistance, and hardware security against rowhammer attacks. He has explored domain-specific languages (Syntheto), incremental modeling techniques (differential-formula), and cloud-edge service resilience against adversarial perturbations. Labs/Teams: Affiliated with the Institute for Software-Integrated Systems (ISIS), focusing on integrating software with physical systems through model-driven approaches and cybersecurity innovations.

• Cyber-physical Systems
• Cybersecurity
• Software Verification
• +۳ مورد دیگر

Daniel Grier is an Assistant Professor jointly appointed in the Computer Science and Engineering and Mathematics departments at the University of California, San Diego (UCSD). His research focuses on quantum complexity theory , particularly exploring near-term quantum computing paradigms and proving quantum advantage over classical systems. He holds a Ph.D. from MIT and was previously a postdoctoral fellow at the University of Waterloo’s Institute for Quantum Computing. Education: Ph.D. in Computer Science, MIT B.S. in Computer Science and Mathematics, University of South Carolina Research Interests: Grier’s work bridges theoretical computer science and quantum computing, emphasizing algorithm design, complexity class separations, and foundational questions about quantum supremacy. He studies how low-depth quantum circuits, boson sampling, and other near-term technologies can achieve computational tasks classically deemed intractable. Recent Article Trends: His publications explore efficient quantum state learning (e.g., classical shadows), hardness results for quantum sampling problems (e.g., bipartite Gaussian boson sampling), and circuit lower bounds (e.g., depth-2 QAC circuits). These contributions highlight his focus on rigorously defining quantum computational advantages. Awards: None explicitly listed in the text. Advising & Grants: Advises at least one student, Jackson Morris. His research is supported by grants exploring quantum complexity and algorithm design. Teaches advanced courses on quantum complexity theory, computability, discrete mathematics, and quantum computing fundamentals. Labs/Teams: Maintains an active lab focused on quantum complexity theory, collaborating with colleagues on topics like interactive protocols and shallow quantum circuits.

• Quantum Complexity Theory
• Quantum Computing
• Computational Complexity
• +۳ مورد دیگر

Danqi Chen is an Associate Professor in the Department of Computer Science at Princeton University's School of Engineering and Applied Science. Their research focuses on advancing large language models (LLMs), with emphasis on model alignment, safety, and long-context reasoning capabilities. Key research areas: LLMs, AI safety, retrieval systems, and model optimization Recent work explores theorem proving, context encoding, and ethical content generation Their 2025 publications highlight innovations in formal verification scaffolding, attention mechanism efficiency, and copyright-aware generation. 2024 studies investigate continual memorization, rule-based chatbot representations, and scientific literature retrieval benchmarks. Current projects demonstrate commitment to improving model robustness, interpretability, and security compliance in multimodal systems.

• Machine Learning
• Natural Language Processing
• Artificial Intelligence
• +۴ مورد دیگر

Dr. Chunyan Mu serves as a Senior Lecturer in the School of Natural and Computing Sciences at the University of Aberdeen, actively contributing to both academic instruction and cutting-edge research in computing science while currently accepting new PhD students. Her research program centers on Trustworthy AI and Safe Autonomy, with specialized expertise in formal verification of responsibility, accountability, and privacy mechanisms within multi-agent systems. She investigates resilience frameworks for autonomous intelligent systems and develops advanced methodologies for information flow security analysis, bridging theoretical computer science with practical security implementations. Analysis of her publication trajectory (2014-2025) reveals consistent innovation in applying formal methods to security-critical systems. Key thematic developments include probabilistic strategy logic for observability analysis, quantitative verification of opacity properties, and game-theoretic approaches to security verification, demonstrating increasing sophistication in handling multi-agent accountability and system resilience challenges. Dr. Mu currently supervises PhD candidates and offers a fully funded doctoral position focused on formal verification of safety properties in autonomous systems, providing comprehensive financial support including tuition coverage, £20,780 annual stipend, and dedicated research funding for candidates with strong backgrounds in formal methods and artificial intelligence.

• Artificial Intelligence
• Computer Security
• Multi-Agent Systems
• +۳ مورد دیگر

Tianyi Zhang is a Tenure-Track Assistant Professor in the Department of Computer Science at Purdue University, part of the College of Science. He leads the Human-Centered Software Systems Lab, focusing on AI-driven systems that synergize human expertise with machine intelligence to enhance programming productivity and software reliability. Prior to Purdue, he was a Postdoctoral Fellow at Harvard University under Dr. Elena Glassman and earned his Ph.D. from UCLA (2019) and B.Sc. from Huazhong University of Science and Technology (2013). Education: Ph.D. in Computer Science, University of California, Los Angeles (2019) Bachelor's in Computer Science, Huazhong University of Science and Technology (2013) Research Interests: His work spans Software Engineering, Human-Computer Interaction, and AI. Key areas include program synthesis, interactive debugging tools, autonomous driving system testing, and mitigating biases in AI models. He develops systems like Interpretable Program Synthesis and SQLucid to bridge human and machine intelligence. Recent Trends in Publications: Recent work emphasizes human-in-the-loop AI, including mixed-initiative systems for data wrangling (Dango), interactive program repair, and bias analysis in text representations (STILE). He also explores challenges in autonomous driving testing and LLM-based code generation errors. Awards & Grants: NSF Career Award (2024) Amazon Research Award Showalter Trust Research Award for pre-diabetes research $1.5M NSF grant for software supply chain security Best Paper Honorable Mentions at CHI and VAHC Advising & Teams: Supervises 12+ PhD/Master's students and 30+ research interns. Notable advisees include Bonan Kou (API misuse studies) and Yuan Tian (text-to-SQL systems). Collaborates with Harvard Medical School on healthcare data analysis. Labs & Initiatives: Directs Purdue's Human-Centered Software Systems Lab. Co-founded the Societal Impact Fellows program. Active in open-source projects like Examplore for API usage visualization and JShrink for Java debloating.

• Artificial Intelligence
• Machine Learning
• Natural Language Processing
• +۴ مورد دیگر

Cezary Kaliszyk is a Professor in Theoretical Computer Science at the University of Melbourne, previously affiliated with the University of Innsbruck. He is actively involved in research and leadership in formal methods, automated reasoning, and machine learning for theorem proving. Research Interests: Automated Reasoning and Interactive Theorem Proving Formalized Mathematics and Proof Automation Machine Learning for Logic and Theorem Proving Integration of AI with Proof Assistants (Coq, Isabelle) Dependent Type Theory and Higher-Order Logic His recent publications (2023–2025) span topics in dependently-typed logic, learning for proof guidance, formalization of surreal numbers, and blockchain-based formal methods. The works consistently bridge formal logic with machine learning, emphasizing automation, explainability, and cross-system integration. Scientific Leadership and Projects: Principal Investigator, ERC project FormalWeb3 Lead Developer, CoqHammer , Tactician , ProofWeb WG5 Leader, COST Action EuroProofNet (until 2024) Contributor to HOL(y)Hammer , Isabelle Enigma He supervises multiple PhD students and has mentored several graduates in formal methods and AI. He teaches courses in theoretical computer science, logic, and machine learning. There are no listed awards in the provided data, but his extensive publication record and project leadership indicate significant recognition in the field. Labs and Research Groups: He leads a research group focused on formal methods and learning-based reasoning, collaborating internationally on projects involving proof automation, formal libraries, and semantic technologies.

• Automated Reasoning
• Formal Methods
• Formalized Mathematics
• +۷ مورد دیگر

Matteo Maffei is a Full Professor at TU Wien, leading the Security and Privacy group. He joined in 2017 after 11 years at Saarland University's CISPA. He holds a Ph.D. in Computer Science from Ca’ Foscari University of Venice (2006). He coordinates the TU Wien Cybersecurity Center, the SecInt Doctoral School, and the FWF Special Research Program SPyCoDe. His research focuses on formal methods for security and privacy, blockchain technologies, and web security. Roles: Full Professor, Coordinator of TU Wien Cybersecurity Center, Module Head of Christian Doppler Lab for Blockchain Technologies (CDL-BOT), Board Member of Vienna Cybersecurity and Privacy Research Cluster (ViSP). His work emphasizes formal verification of cryptographic protocols, smart contracts, and decentralized systems. Key achievements include ERC Advanced (2024) and Consolidator (2018) Grants, and leadership roles in conferences like IEEE Computer Security Foundations Symposium (CSF). Research Interests: Formal methods, smart contracts, blockchain scalability, web security, privacy-preserving protocols, and decentralized systems. His recent projects include optimizing Lightning Network channels, secure multi-hop payments, and AI-driven robustness verification. Publications: Over 200 publications in top venues like CCS, IEEE S&P, and CRYPTO. Recent work includes advancements in blockchain interoperability (e.g., Alba bridges), light client protocols (Blink), and neural network verification techniques. Grants & Awards: ERC Advanced Grant (2024), ERC Consolidator Grant (2018), DFG Emmy Noether Fellowship (2009). Led projects funded by EU Horizon, FWF, and industry partners like ABC Research GmbH. Advising & Labs: Supervised over 20 PhD/Master theses. Active in the Christian Doppler Lab for Blockchain Technologies and the SPyCoDe SFB. Collaborates with institutions like SBA Research and Stanford University.

• Computer Security
• Cryptography
• Privacy
• +۴ مورد دیگر

Floris van Doorn is a Professor at the Mathematical Institute of the University of Bonn where he leads the Formalized Mathematics group. His research focuses on making it viable to formalize research mathematics in proof assistants that can check the correctness of such proofs. He primarily works with the Lean Theorem Prover and is a maintainer of its mathematical library (mathlib). University of Bonn: Professor (2023-present) University of Paris-Saclay: Postdoc with Patrick Massot (2021-2023) University of Pittsburgh: Postdoc with Tom Hales (2018-2021) Carnegie Mellon University: PhD under Jeremy Avigad and Steve Awodey (2013-2018) Van Doorn's research interests center on formalized mathematics, tools and automation for formalization, and homotopy type theory. He has made significant contributions to several major formalization projects including the Carleson project (proving Carleson's theorem), the sphere eversion project (formalizing Gromov's h-principle), the Flypitch project (formalizing the independence of the continuum hypothesis), and the Spectral sequences project. His work demonstrates that proof assistants can handle complex areas of mathematics beyond algebra, including differential topology and analysis. His recent publications show a consistent focus on advancing formalized mathematics, with his most recent work formalizing the Gagliardo-Nirenberg-Sobolev inequality and continuing the Carleson project. His publications span theoretical foundations of type theory, practical applications of formalization, and educational resources for learning proof assistants. Skolem award (2025) for the paper 'The Lean Theorem Prover (System Description)' Van Doorn actively mentors students and collaborators, with Maria, Michael, and Arend recently joining his formalization group in Bonn. He has taught various courses on formalized mathematics and proof assistants at the University of Bonn, University of Pittsburgh, and Carnegie Mellon University. His educational efforts include developing learning resources such as the Natural Number Game and the online book 'Mathematics in Lean.' He also maintains an active presence in the Lean community through the Formalized Mathematics group and collaborative projects like the Carleson project, which invites participation from those familiar with Lean.

• Formalized Mathematics
• Proof Assistants
• Homotopy Type Theory
• +۳ مورد دیگر