News
Two papers accepted to ICLR 2026
Two papers from our group accepted to ICLR 2026 in Rio de Janeiro, Brazil.
Paper accepted to EACL 2026
LIAR: Leveraging Inference Time Alignment (Best-of-N) to Jailbreak LLMs in Seconds
Paper on safety of VLMs accepted to AAAI 2026
Three papers accepted to NeurIPS 2025
Work spanning reasoning models, theoretical RL, and LLM alignment. San Diego, CA.
Paper accepted to EMNLP 2025
Paper accepted to CDC 2025
Three papers accepted to IROS 2025
Paper accepted to ICML 2025
Vancouver, Canada.
Paper accepted to IEEE RAL 2025
Multi-robot coordination via locality-based factorized multi-agent actor-critic.
Paper accepted to CVPR 2025
Immune: Improving Safety Against Jailbreaks in Multi-modal LLMs via Inference-Time Alignment
Organizing the first NeurIPS Workshop on Biosecurity Safeguards for Generative AI
San Diego, California, USA.
Invited talk at Asilomar 2025 on satisficing alignment
Presented Transfer Q* at UT Austin
Area Chair — ACL 2025, NeurIPS 2025
Area Chair — Transactions on Machine Learning Research (TMLR)
Two papers accepted to NeurIPS 2024
Vancouver, Canada.
Five papers accepted to ICML 2024
Vienna, Austria. Including one Spotlight (top 3.5% of submissions).
Three papers accepted to IROS 2024
Abu Dhabi, UAE.
Paper accepted to ICLR 2024
PARL: A Unified Framework for Policy Alignment in Reinforcement Learning
Paper accepted to JMLR
On the Sample Complexity and Metastability of Heavy-tailed Policy Search in Continuous Control
Paper accepted to TMLR
A Survey on the Possibilities & Impossibilities of AI-Generated Text Detection
Invited talks at IBM Research, Google DeepMind NYC, and Amazon
Presented Transfer Q* and MaxMin-RLHF. Also presented at Amazon Rising Star Speaker Series (June 2024).
Area Chair — NeurIPS 2024
Paper accepted (oral) at CoRL 2023
iPLAN: Intent-Aware Planning in Heterogeneous Traffic via Distributed Multi-Agent RL — also won Best Paper at the Advances in Multi-Agent Learning Workshop at IROS 2023.
Two papers accepted to ICML 2023
STEERING and Beyond Exponentially Fast Mixing in Average-Reward RL via Multi-Level Monte Carlo Actor-Critic.
Three papers accepted to ICRA 2023
RTAW, Decentralized Multi-agent Exploration, and Heavy-Tailed Policy Optimization for sparse rewards.
Two papers accepted (oral) to AAAI 2023
Paper accepted to ICLR 2023
SWIFT: Rapid Decentralized Federated Learning via Wait-Free Model Communication
Paper accepted to JAIR
Achieving Zero Constraint Violation for Constrained Reinforcement Learning via Primal-Dual Approach
Paper accepted to IEEE TAI
Regret and Belief Complexity Trade-off in Gaussian Process Bandits via Information Thresholding
Area Chair — NeurIPS 2023
Two papers accepted to ICML 2022 (spotlight talks)
On the Hidden Biases of Policy Mirror Ascent; FedNew for federated learning.
Two papers accepted to IROS 2022
Distributed Riemannian Optimization; DC-MRTA for multi-robot task allocation.
Paper accepted to CoRL 2022
HTRON: Efficient Outdoor Navigation with Sparse Rewards via Heavy Tailed Adaptive Reinforce.
Paper accepted to IEEE TSP
Projection-Free Algorithm for Stochastic Bi-level Optimization.
Two papers accepted to AAAI 2022
Amazon Research Award 2022
For the proposal "Ensuring fairness via federated learning beyond consensus" at the University of Maryland.
ARL Grant — Efficient Multi-Agent RL for Next-generation Team of Combat Vehicles
Principal Investigator.
ARL Grant — Risk Sensitive Maneuver Tradeoffs for a Team of Ground Vehicles
Co-Principal Investigator.
Northrop Grumman Seed Grant Award
Principal Investigator.