Radha Poovendran
Professor
Studies security of networked systems and the safety of large language models, including jailbreak attacks and defenses.
Radha Poovendran's lab at UW works on trustworthy & safe ML and large language models. 55 papers since 2023, 19 at top venues, mostly CDC and ACL. 7 PhD students and 1 master's student. 41 alumni, most often to Amazon and Boeing Research & Technology. Applications close .
At a glance
- papers, 2023–2026
- 14/ year
- at T1 venues
- 35%
- PhD students
- 7
- MS students
- 1
- alumni listed
- 41
Papers by yearfrom the lab homepage and DBLP, 2023–2026
- 5 papers2026 · T1 3 · collecting
- 12 papers2025 · T1 4
- 22 papers2024 · T1 9
- 16 papers2023 · T1 3
T1 venuesOtherVenues: CDC, ACL, Findings of ACL, ICLR, NAACL
Research focuspaper titles that use each topic's words
- Large language models18 papers
- attacks7 papers
- Trustworthy & safe ML2 papers
- studies security0 papers
- defenses0 papers
Paper list
From the lab homepage and DBLP.
20265 papers
- ACL (T1)BadScientist: Can a Research Agent Write Convincing but Unsound Papers that Fool LLM Reviewers? (opens in a new tab)
- ACM Trans. Cyber Phys. Syst.Electric Vehicles Security and Privacy: Challenges, Solutions, and Future Needs (opens in a new tab)
- ICLR (T1)SoSBench: Benchmarking Safety Alignment on Six Scientific Domains (opens in a new tab)
- ACL (T1)Temporal Sampling for Forgotten Reasoning in LLMs (opens in a new tab)
- Trans. Mach. Learn. Res.TinyV: Reducing False Negatives in Verification Improves RL for LLM Reasoning (opens in a new tab)
202512 papers
- IEEE Trans. Dependable Secur. Comput.CANLP: Intrusion Detection for Controller Area Networks Using Natural Language Processing and Embedded Machine Learning (opens in a new tab)
- ACNSCANTXSec: A Deterministic Intrusion Detection and Prevention System for CAN Bus Monitoring ECU Activations
- AAAI (T1)ChatBug: A Common Vulnerability of Aligned LLMs Induced by Chat Templates (opens in a new tab)
- IEEE Trans. Dependable Secur. Comput.Identity-Based Authentication for On-Demand Charging of Electric Vehicles (opens in a new tab)
- Findings of ACLKodCode: A Diverse, Challenging, and Verifiable Synthetic Dataset for Coding (opens in a new tab)
- ICLR (T1)Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing (opens in a new tab)
- Findings of ACLSafeChain: Safety of Language Models with Long Chain-of-Thought Reasoning Capabilities (opens in a new tab)
- Findings of ACLSmall Models Struggle to Learn from Strong Reasoners (opens in a new tab)
- NAACL (T1)Stronger Models are Not Always Stronger Teachers for Instruction Tuning (opens in a new tab)
- —Stronger Models are NOT Stronger Teachers for Instruction Tuning (opens in a new tab)
- CDCSubmodular Swarm Assignment for Multi-Agent Systems under Linear Temporal Logic Constraints
- ICML (T1)ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning (opens in a new tab)
202422 papers
- —''A Hybrid Submodular Optimization Approach to Controlled Islanding with Post-Disturbance Stability Guarantees,'' (opens in a new tab)
- —"A Principled Incentive Mechanism to Promote Economic Viability of Mini-Grids" (opens in a new tab)
- ACL (T1)"ArtPrompt: ASCII Art-based Jailbreak Attacks against Aligned LLMs" (opens in a new tab)
- ICLR (T1)"BadChain: Backdoor Chain-of-Thought Prompting for Large Language Models" (opens in a new tab)
- TAC (T1)“Dynamic Information Flow Tracking for Detection of Advanced Persistent Threats: A Stochastic Game Approach” (opens in a new tab)
- AsiaCCS"POSTER: Identifying and Mitigating Vulnerabilities in LLM-Integrated Applications" (opens in a new tab)
- ACC"Risk-Aware Distributed Multi-Agent Reinforcement Learning" (opens in a new tab)
- TAC (T1)“RL-ARNE: A Reinforcement Learning Algorithm for Computing Average Reward Nash Equilibrium of Nonzero-Sum Stochastic Games” (opens in a new tab)
- ACL (T1)"SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding" (opens in a new tab)
- Autom.“Stochastic Dynamic Information Flow Tracking Game using Supervised Learning for Detecting Advanced Persistent Threats” (opens in a new tab)
- USENIX Security (T1)ACE: A Model Poisoning Attack on Contribution Evaluation Methods in Federated Learning (opens in a new tab)
- SACCANLP: NLP-Based Intrusion Detection System for CAN" (opens in a new tab)
202316 papers
- CCS (T1)''MDTD: A Multi-Domain Trojan Detector for Deep Neural Networks,'' (opens in a new tab)
- CDC"A Compositional Resilience Index for Computationally Efficient Safety Analysis of Interconnected Systems" (opens in a new tab)
- IEEE Trans. Inf. Forensics Secur."A Natural Language Processing Approach for Instruction Set Architecture Identification." (opens in a new tab)
- CDC"A Submodular Energy Function Approach to Controlled Islanding with Provable Stability" (opens in a new tab)
- ACM Trans. Cyber Phys. Syst."A Timing-Based Framework for Designing Resilient Cyber Physical Systems under Safety Constraint," (opens in a new tab)
- AsiaCCS"LDL: A Defense for Label-Based Membership Inference Attacks,'' (opens in a new tab)
- IJCAI (T1)"Learning Dissemination Strategies for External Sources in Opinion Dynamic Models with Cognitive Biases," (opens in a new tab)
- CDC"Necessary and Sufficient Conditions for Satisfying Linear Temporal Logic Constraints using Control Barrier Certificates" (opens in a new tab)
- 2, April"Robust Satisfaction of Metric Interval Temporal Logic Objectives in Adversarial Environments," (opens in a new tab)
- —A Common Framework for Resilient and Safe Cyber-Physical System Design
- WISECBARON: Base-Station Authentication Through Core Network for Mobility Management in 5G Networks (opens in a new tab)
- International Conference on Autonomous Agents and Multiagent Systems (AAMAS ), MayCognitive Bias-Aware Dissemination Strategies for Opinion Dynamics with External Information Sources
Where alumni went41 listed, 45% to industry
Industry 5Academia 6
Most often to
- Amazon
- Boeing Research & Technology
- Hyundai Motor Company
Also to
- California State University Los Angeles Web Page maruf3e
- Embry-Riddle Aeronautical University
- Iowa State University Web Page LinkedIn
Apply to UWFall 2027
UW Allen School Computer Science PhD
- Round
- Single round
- Opens
- GRE
- not required
- English
- See program page
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- USUSC
Bart Andrew Kosko
0 T1 - CNTsinghua
Bo Zhang
- Learning theory & optimization
- CHETH Zurich
Martin Vechev
- Trustworthy & safe ML
- GBImperial
Danilo Mandic
- Learning theory & optimization
- GBCambridge
Ferenc Huszár
7 T1 - HKCUHK
Weiyang Liu
- Generative & foundation models
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage and DBLP
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
UWcloses