جستجوی استاد راهنمای دانشگاهی

Supratik Chakrabortyمشاهده پروفایل

استاد

• Formal Methods
• Formal Verification
• Automated Synthesis
• +۷ مورد دیگر

Supratik Chakraborty serves as the Bajaj Group Chair Professor in the Department of Computer Science and Engineering at Indian Institute of Technology Bombay. He maintains dual affiliations with the Centre for Formal Design and Verification of Software and the Centre for Liberal Education at IIT Bombay, demonstrating his cross-disciplinary engagement. Professor Chakraborty's research spans formal methods with focus on formal verification, rigorous analysis of system models, and automated synthesis of systems from specifications. His work bridges theoretical foundations with practical applications, particularly in developing mathematically provable guarantees for increasingly complex hardware, software, and intelligent systems. Current research interests include constrained counting and sampling, scalable formal verification of software and hardware systems, automated synthesis of programs and circuits, and applications of automata, logic and finite model theory to practical verification challenges. His publication trajectory shows a significant evolution from traditional hardware and software verification toward addressing verification challenges in machine learning and AI systems. Recent work increasingly focuses on interpretability of black-box models, verification of neural networks, and synthesis techniques applicable to intelligent systems. The research demonstrates strong interdisciplinary connections between formal methods, programming languages, and artificial intelligence. IIT Bombay Excellence in Thesis (CSE) Award 2011 (for Bhargav Gulavani's thesis) IIT Bombay Excellence in Thesis (CSE) Award 2017 (for Abhisekh Sankaran's thesis) Best Paper in Algorithms and Architecture track at IEEE International Conference on Computer Design: VLSI in Computers and Processors, 1998 Professor Chakraborty has successfully supervised 11 doctoral students, with research spanning formal verification techniques, Boolean functional synthesis, constrained counting, and applications to hardware and software systems. His students have gone on to positions at major institutions including Microsoft Research, TCS Research, Georgia Tech, and BARC, reflecting the strong industry and academic impact of his mentorship. Current research directions show increasing emphasis on verification challenges posed by machine learning systems and AI. His research group at IIT Bombay, while not explicitly named in the materials, appears to focus on formal methods with strong connections to the Centre for Formal Design and Verification of Software. The group maintains active collaborations with international researchers including Moshe Y. Vardi at Rice University, and has made significant contributions to verification tools like VeriAbs that bridge theoretical advances with practical applications.

Sean Welleckمشاهده پروفایل

استادیار

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

Caroline Lemieuxمشاهده پروفایل

استادیار

• Programming Languages
• Software Engineering
• Fuzz Testing
• +۶ مورد دیگر

Caroline Lemieux is an Assistant Professor at the Department of Computer Science, University of British Columbia (UBC), with research focused on advancing software correctness, security, and performance through innovative testing and synthesis techniques. Her work bridges Programming Languages and Software Engineering , particularly in fuzz testing, specification mining, and program synthesis. PhD from University of California, Berkeley (2021), advised by Koushik Sen Postdoctoral researcher at Microsoft Research, NYC (2021-2022) Key contributions: FuzzFactory , CodaMOSA , Arvada , and Gauss Her research integrates machine learning with traditional testing methods, exemplified by projects like RLCheck (reinforcement learning for test generation) and AutoPandas (neural synthesis for dataframes). Recent publications analyze generator-based fuzzing challenges and propose hybrid strategies combining coverage feedback with AI-driven insights. Scientific Awards : ACM/SIGSOFT Best Paper Award (ESEC/FSE 2019) ACM/SIGSOFT Tool Demonstration Award (ISSTA 2019) ACM/SIGSOFT Distinguished Artifact Award (ISSTA 2019) NSERC Postgraduate Scholarship-Doctoral (PGS D) UBC Governor General's Silver Medal (2016) Teaching roles include: 2025W2: CPSC 539L - Topics in Programming Languages 2024W2: CPSC 410 - Advanced Software Engineering 2023W2: CPSC 410 (with Alex Summers) She supervises graduate and undergraduate researchers working on projects like ExploTest (automated unit test generation) and GRIMOIRE (grammar extraction from pseudo-rules). Her research team collaborates with institutions including Microsoft Research, Google, and academic partners in systems security and AI-driven testing.

Ardalan Amiri Saniمشاهده پروفایل

دانشیار

• Mobile Systems
• Operating Systems
• Security
• +۵ مورد دیگر

Ardalan Amiri Sani is an Associate Professor in the Computer Science Department at the University of California, Irvine (UCI), affiliated with the Donald Bren School of Information & Computer Sciences. He leads the Trustworthy Systems Lab (TrussLab), focusing on secure and reliable mobile systems, operating systems, and virtualization. His research bridges mobile computing, security, and OS design, addressing challenges in I/O devices, kernel hardening, and verifiable provenance. Education: Ph.D. and M.Sc. from Rice University (ECE), B.Sc. from Sharif University of Technology. Awards include the NSF CAREER Award, Google ASPIRE Award, and UCI Dean's Mid-Career Research Award. He has advised numerous graduate and undergraduate students, many of whom now work in academia and industry. Research highlights include Tabellion (secure legal contracts on mobile devices), ProvCam (verifiable video provenance), and work on minimizing smartphone TCBs via hardware isolation. His grants include NSF funding for OS kernel security and NSA support for provenance systems. Education: Ph.D., Electrical and Computer Engineering, Rice University, 2014 B.Sc., Electrical Engineering, Sharif University of Technology Grants: $500K NSF Award (with UCR) for OS kernel security (2020) NSF SaTC Award for GPU security in browsers Google ASPIRE Award for Android system call filtering Labs/Teams: Trustworthy Systems Lab (TrussLab), collaborating with industry partners like Intel and Broadcom.

Juan Zhaiمشاهده پروفایل

استادیار

• Software Engineering
• Program Languages
• Natural Language Processing
• +۶ مورد دیگر

Juan Zhai is an Assistant Professor in the Manning College of Information and Computer Sciences (CICS) at the University of Massachusetts Amherst. She co-directs the Laboratory for Advanced Software Engineering Research (LASER) and is a member of the UMass NLP group. Her research advances software engineering through automated techniques for building high-quality systems with emphasis on behavioral specifications, AI safety, and trustworthy AI. Her work addresses the fundamental challenge of aligning software behavior with intended specifications through two main directions: automated specification synthesis (translating natural language comments to formal specifications via tools like C2S and LLMCup) and defect detection/repair (developing frameworks for AI system testing, bias mitigation, and training diagnostics). Her vision integrates these into end-to-end assurance systems that continuously validate, repair, and audit evolving software in dynamic environments. Recent publications (2024-2025) reveal dominant trends at the software engineering/AI intersection: formal specification synthesis for IoT and code generation, comment maintenance using LLMs, deep learning framework testing (DevMuT, Citadel), bias detection in LLMs, and automated training repair (AutoTrainer, DREAM). These contributions appear in top venues including ICSE, FSE, ASE, ISSTA, and ACL. Professor Zhai currently advises PhD student Gehao Zhang (focusing on Software Engineering and AI Safety) and actively recruits new PhD/Master's students. Her LASER lab develops practical tools for specification inference, LLM-driven synthesis, and trustworthy AI, while collaborating with the UMass NLP group on language-centric software analysis. The LASER lab, co-directed by Zhai, pioneers techniques for behavioral specification enforcement across traditional and AI-powered systems. Key projects include CPC for bidirectional code-comment analysis, ModelMeta for deep learning framework testing, and frameworks for bias mitigation across the ML lifecycle. The lab emphasizes practical, scalable tools that enhance correctness, robustness, and fairness in critical AI applications.

Daniel Varroمشاهده پروفایل

استاد

• Model-Based Systems Engineering
• Cyber-Physical Systems
• Software Engineering
• +۱۲ مورد دیگر

Daniel Varro is a Professor affiliated with McGill University (Faculty of Engineering, School of Computer Science), with strong ties to Budapest University of Technology and Economics and Linköping University. He is a leading researcher in model-driven engineering, cyber-physical systems, and software engineering, actively contributing to top-tier conferences such as MODELS, ICSE, and ASE. His research focuses on model-based systems engineering (MBSE) , automated model generation , model transformations , and constraint-based consistency checking . Recently, his work has expanded into integrating large language models (LLMs) and machine learning into modeling workflows, including model querying, domain modeling, and bug detection. The recent publications reveal a strong trend toward AI-augmented modeling, logic-based solvers (e.g., Refinery), and safety assurance of autonomous systems (e.g., COLREGs compliance). His work bridges formal methods with practical software engineering challenges in industrial and safety-critical domains. Scientific Awards: No specific awards mentioned in the text. Advising and Grants: While no explicit list of students or grants is provided, his mentorship in the Doctoral Symposium and repeated leadership roles suggest active supervision and likely grant funding. He has led projects on automated model generation, model quality, and AI integration in modeling. Labs and Teams: Daniel Varro is associated with research groups focused on model-driven engineering and software evolution, likely leading or co-leading teams working on the VIATRA and Refinery frameworks for model transformation and solving.

• Cybersecurity
• Software Security
• System Security
• +۶ مورد دیگر

Ruoyu (Fish) Wang is an Associate Professor at the School of Computing and Augmented Intelligence, Arizona State University (Tempe campus). He also holds affiliations as Associate Director of Impact at the Global Security Initiative, Center for Cybersecurity & Trusted Foundations, and with the Biodesign Center for Biocomputing, Security and Society. His educational background includes: Ph.D. in Computer Science, University of California, Santa Barbara Professor Wang's research focuses on system security, with an emphasis on automated binary program analysis and reverse engineering of software. He is the co-founder and core developer of the angr binary analysis platform, which won third place in the DARPA Cyber Grand Challenge (2018). His work spans vulnerability discovery, fuzzing, and security tool development for binary program analysis. His current research interests include: Binary program analysis and reverse engineering Automated vulnerability discovery and mitigation Fuzzing techniques and robust testing Phishing and fraud detection in e-commerce Security of firmware and embedded systems Application of machine learning to security problems His recent publications (2024-2025) demonstrate cutting-edge research in fraud detection for e-commerce using LLMs, advanced fuzzing methodologies, and binary decompilation techniques. Key trends include bridging theoretical program analysis with practical security tools, as evidenced by extensions to the angr platform, and addressing emerging threats in financial ecosystems and client-side security. Dr. Wang has received notable recognition: Third place in DARPA Cyber Grand Challenge (2018) with team Shellphish As an active educator, he supervises graduate research (CSE 599/799) and teaches core cybersecurity courses including Software Security (CSE 545) and Information Assurance (CSE 365). His teaching spans multiple semesters through 2025, covering practicums, internships, and special topics in computing security. Dr. Wang co-founded the angr binary analysis platform and contributes to Arizona State University's security research ecosystem through leadership roles in the Center for Cybersecurity & Trusted Foundations and Biodesign Center for Biocomputing, Security and Society.

Stefania Dumbravaمشاهده پروفایل

دانشیار

• Formal methods for databases
• Graph databases
• Property graphs
• +۷ مورد دیگر

Stefania Dumbrava is an Associate Professor of Computer Science at ENSIIE (École Nationale Supérieure d'Informatique pour l'Industrie et l'Entreprise) and a permanent member of the ACMES team in the SAMOVAR laboratory at Télécom SudParis, Institut Polytechnique de Paris. She is also actively involved in the Property Graph Schema Working Group and the European Research Network on Formal Proofs (EuroProofNet). Education PhD in Computer Science, Université Paris-Sud (2016) MSc in Computer Science, Jacobs University Bremen (2012) BSc in Mathematics, Jacobs University Bremen (2010) Research Interests Dumbrava's research lies at the intersection of formal methods and data management . She designs and verifies algorithms and systems for graph databases , with emphasis on property graphs , schema discovery , query optimization , and distributed graph processing . Recently, her work focuses on certifying large-scale distributed graph systems under the ANR JCJC VERDI project (2025–2029). Awards & Honors SIGMOD Best Paper Award 2023 – “PG-Schema: Schemas for Property Graphs” SIGMOD Research Highlight Award 2023 – “Threshold Queries” VLDB 2022 Best Regular Research Paper Runner-Up – “Threshold Queries in Theory and in the Wild” SIGMOD 2025 Distinguished Reviewer Award ICDE 2025 Best Program Committee Member Award EASST Best Software Science Paper Award, ICGT 2025 Students & Grants Dumbrava has supervised numerous research interns and is actively recruiting PhD students for her ANR VERDI project on verified foundations of large-scale distributed graph systems. She has also served on six PhD thesis committees as examiner since 2021. Labs & Teams She leads the ACMES research group within the SAMOVAR laboratory (Télécom SudParis, Institut Polytechnique de Paris), where her team develops formally verified graph-database engines and tools such as GRASP, VerDILog, and DatalogCert.

Debjit Palمشاهده پروفایل

پژوهشگر ارشد

• Machine Learning for Electronic Design Automation (EDA)
• System-on-Chip (SoC) Verification
• Hardware Verification
• +۶ مورد دیگر

Debjit Pal is a Post-Doctoral Associate at the School of Electrical and Computer Engineering, Cornell University, and a member of the Computer Systems Laboratory. His research focuses on machine learning techniques for hardware verification, SoC validation, and FPGA optimization. Education: Ph.D. in Computer Engineering (University of Illinois at Urbana-Champaign, 2019) M.S. in Computer Science (IIT Kharagpur, 2012) B.E. in Electronics Engineering (Jadavpur University, 2008) Research Interests: Machine Learning for Electronic Design Automation (EDA) System-on-Chip (SoC) Verification Edge Intelligence as a Service Compiler Optimizations for Reconfigurable and High-Performance Computing Scientific Awards: IEEE CEDA Student Research Award (2016) Best Paper Nomination (ICCAD 2015, DAC 2018, ASP-DAC 2019) E. J. McCluskey Best Doctoral Thesis Competition Semi-Finalist (2020) Travel Grants for ICCAD/DAC/ASPDAC (2018-2019) Professional Roles: Technical Program Committee Member (DAC, VLSID), Reviewer (IEEE TVLSI, DATE, ICCAD). Collaborates with researchers like Zhiru Zhang and Shobha Vasudevan.

• Network Security
• Android Security
• Malware Analysis
• +۶ مورد دیگر

Vijay Laxmi is a Professor in the Department of Computer Science and Engineering at Malaviya National Institute of Technology (MNIT) Jaipur, India. With over a decade of active research publication from 2014-2024, Dr. Laxmi has established themselves as a prominent researcher in network security, Android security systems, and Network-on-Chip architectures. Their work demonstrates consistent collaboration with Manoj Singh Gaur and numerous doctoral students at MNIT Jaipur. Dr. Laxmi's research interests span Network Security, Android Security, Malware Analysis, Network-on-Chip Architectures, Routing Protocols, Side-Channel Attacks, Wireless Networks, and Mobile Security. Their work bridges theoretical security frameworks with practical implementations, particularly in mobile and embedded systems. Recent publications indicate a growing focus on AI-based security approaches including GAN applications for fuzzing and deep learning for image dehazing. The research trajectory shows increasing sophistication in security analysis techniques, evolving from basic malware detection to advanced side-channel attack analysis and sophisticated network security protocols. Recent publications demonstrate expertise in both theoretical frameworks and practical implementations with applications in real-world security challenges. Dr. Laxmi has mentored numerous graduate students including Vineeta Jain, Anugrah Jain, Sonal Yadav, Mohit Singh, and Gaurav Singal, who appear as co-authors across multiple publications. Their collaborative network extends to researchers at international institutions, indicating strong academic connections beyond their home institution.

Wolfgang Kunzمشاهده پروفایل

استاد

• Hardware Safety and Security
• Formal Hardware Verification
• Embedded Computing Systems
• +۵ مورد دیگر

Wolfgang Kunz is a Full Professor (C4, W3) and Chair of Electronic Design Automation at the Technische Universität Kaiserslautern since 2001. His academic career spans multiple prestigious institutions, including Goethe-University Frankfurt/Main and the University of Massachusetts, Amherst. He has held leadership roles such as Dean (2005-2007) and Vice-Dean (2007-2009) at TU Kaiserslautern. Habilitation (Dr. rer. nat. habil.), Computer Science, University of Potsdam (1996) Doctoral degree (Dr.-Ing.), Electrical Engineering, University of Hannover (1992) Dipl.-Ing. degree, Karlsruhe Institute of Technology (1989) His research focuses on hardware verification, security, and optimization, particularly in embedded systems and processors. His work on formal verification methods has been commercialized by companies like Synopsys, Mentor Graphics, and Siemens EDA. His 2016-2021 publications address critical security issues such as Spectre/Meltdown and introduce innovative verification frameworks adopted by industry leaders like Infineon and OneSpin Solutions. Scientific awards include the IEEE Fellow (2006), German IT Society Award (2005), and TU Kaiserslautern Distinguished Teaching Award (2016). He has served on editorial boards of major journals and coordinated the Erasmus Mundus European Master Program in Embedded Computing Systems since 2010. Key students: Jörg Bormann, Raik Brinkmann, Tobias Ludwig Collaborations: Siemens EDA, Infineon, AbsInt, Intel SCAP Spin-offs: LUBIS EDA, OneSpin Solutions

Remko Van Hoekمشاهده پروفایل

استاد

• Supply Chain Management
• Procurement
• Sourcing
• +۵ مورد دیگر

Remko Van Hoek serves as Professor of Practice in the Department of Supply Chain Management at the Sam M. Walton College of Business, University of Arkansas. He teaches Sourcing and Procurement courses across undergraduate, master's, and doctoral programs. Prior to joining the University of Arkansas, Dr. Van Hoek held academic positions in Europe and served as a visiting professor at Cranfield School of Management, complemented by extensive industry experience as a supply chain and procurement executive at global corporations including Nike, PwC, and The Walt Disney Company. He currently serves on the Council of Supply Chain Management Professionals (CSCMP) Board of Directors and acts as executive director of the CSCMP Supply Chain Hall of Fame hosted by the Walton College. Dr. Van Hoek's research centers on digital transformation in procurement and supply chain management, with particular focus on blockchain implementation, artificial intelligence applications, and sustainable sourcing practices. His work investigates how emerging technologies reshape procurement processes, supplier relationship management, and risk mitigation strategies. He explores innovative approaches to supplier diversity programs, ethical sourcing frameworks, and the strategic evolution of procurement from cost reduction to value creation. His research consistently bridges academic theory with industry practice through case studies from major corporations, developing actionable frameworks for practitioners facing digital disruption and sustainability challenges. Analysis of his 15 most recent publications (2023-2025) reveals dominant themes in AI-driven supply chain risk prevention, blockchain implementation for transparency, and sustainable supplier engagement. His work demonstrates consistent industry collaboration, drawing from case studies at Walmart, Moet Hennessy, and Bayer to address post-pandemic resilience challenges. A significant portion examines procurement's strategic evolution beyond transactional functions, with recurring emphasis on ethical considerations in digital transformation and the practical implementation barriers for emerging technologies in global supply networks. As a Professor of Practice, Dr. Van Hoek integrates executive-level industry experience into curriculum development and student mentorship. His industry-engaged teaching model includes guest lecturer programs connecting students with supply chain practitioners, as evidenced by his co-authored work on integrating industry insights into supply chain education. While specific grant funding details are not publicly documented, his leadership in the CSCMP Supply Chain Hall of Fame demonstrates commitment to professional development and industry-academia knowledge transfer. Dr. Van Hoek directs the CSCMP Supply Chain Hall of Fame initiative, which documents transformative contributions to supply chain management through interviews with industry pioneers and historical case studies. This platform serves as both an educational resource for Walton College students and a professional development tool for supply chain practitioners globally. The Hall of Fame preserves critical industry knowledge while highlighting contemporary innovations in supply chain strategy and technology implementation.

• Cryptography
• Information Security
• Cryptographic Automation
• +۳ مورد دیگر

Christina L. Garman is an Assistant Professor in the Department of Computer Science at Purdue University, where she joined in Spring 2018. Her research focuses on practical cryptography and cryptographic automation to make secure system development accessible to non-experts through error-resistant design methodologies. Her educational background includes: Bachelor of Science in Computer Science and Engineering from Bucknell University (2011) Bachelor of Arts in Mathematics from Bucknell University (2011) Master of Science in Engineering in Computer Science from Johns Hopkins University (2013) Doctor of Philosophy in Computer Science from Johns Hopkins University (2017) Professor Garman's work centers on real-world cryptographic system security, spanning protocol analysis (e.g., RC4 in TLS, Apple iMessage flaws), decentralized anonymous systems (Zerocash/ZCash), and cryptographic automation. She pioneered techniques for removing human error in cryptographic deployments through automated tools and frameworks. Her research bridges theoretical cryptography with practical implementation challenges in privacy-preserving technologies and secure infrastructure. Analysis of her 2021-2025 publications reveals expanding research horizons: hardware security vulnerabilities (Rowhammer, SGX), privacy network enhancements (Tor onion services), software supply chain security (SBOM tools), and advanced cryptographic protocols (zkSNARKs, MPC). This evolution demonstrates consistent focus on real-world security impact while diversifying into hardware-software cross-layer threats and formal verification methods for cryptographic implementations. Her major scientific recognitions include: NSF CAREER Award (2021) for cryptographic automation research ACM CCS Best Paper Award (2016) for iMessage security analysis IEEE Test of Time Award (2024) for foundational Zerocash work Professor Garman co-founded ZCash, a privacy-focused cryptocurrency based on her Zerocash protocol, and her NSF CAREER grant supports cryptographic automation development. Her research has received significant media coverage in The Washington Post, Wired, and The New York Times, highlighting real-world relevance. While specific student advising details aren't public, her active publication record indicates ongoing mentorship of graduate researchers in security and cryptography. Her work maintains strong industry connections through ZCash development and Tor network contributions, with recent projects like keyless CDNs demonstrating practical applications of cryptographic automation. She remains a leading voice in cryptographic research communities through conference participation and collaborative projects addressing evolving security challenges.

• Artificial Intelligence
• Computer Security
• Multi-Agent Systems
• +۳ 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.

Tatsunori Hashimotoمشاهده پروفایل

استادیار

• Artificial Intelligence
• Machine Learning
• Natural Language Processing
• +۳ مورد دیگر

Tatsunori Hashimoto is an Assistant Professor of Computer Science at Stanford University, specializing in artificial intelligence, machine learning, and natural language processing. His research focuses on developing robust and ethical language models, addressing challenges in bias mitigation, fairness, and transparency. He leads projects like the Stanford Alpaca, exploring instruction-following models and their societal impacts. Key research interests include generative models, AI ethics, and privacy-preserving techniques. His recent work examines language model behaviors, security risks, and the societal implications of AI systems. Notable contributions include frameworks for auditing language models, improving factual accuracy, and reducing disparities in speech recognition. His publications highlight advancements in long-context processing, few-shot learning, and automated benchmarking. He emphasizes practical applications of AI while addressing dual-use concerns and ensuring alignment with human values.