Sumit Kumar Jha
Eminent Scholar Chair Professor of Computer Science
384 CASE Building
11200 SW 8th Street
Miami, FL 33199
Prof. Jha’s research interests lie in the following areas:
- Artificial Intelligence
- Logic and Automata Theory
- Quantum Computing
- Emerging Architectures
Principal Investigator for $17 million in awarded projects since 2014, including DARPA GARD, AFRL ML-RCP, ONR Science of AI, Department of Energy EXPRESS, NSF Scalable Parallelism in the Extreme (SPX), NSF Software and Hardware Foundations (SHF), NSF Formal Methods in the Field (FMitF), NSF Secure and Trustworthy Computing (SaTC), National Nuclear Security Administration (NNSA), Department of Energy (DOE) ASCR RENEW, and AFOSR Young Investigator Program.
news
Sep 27, 2024 | DAIR.AI listed our work on jailbreaking LLMs as one of the top 10 ML Papers of the Week! |
---|---|
Feb 11, 2024 | Three (3) papers have been accepted at the ACM Design Automation Conference (DAC) 2024! |
selected publications
- Integrated Decision Gradients: Compute Your Attributions Where the Model Makes Its DecisionProceedings of the 38th AAAI International Conference on Artificial Intelligence (AAAI), 2024
- Jailbreaking Large Language Models with Symbolic MathematicsWorkshop on Socially Responsible Language Modelling Research (SoLaR) at NeurIPS, 2024
- AutoSafeCoder: A Multi-Agent Framework for Securing LLM Code Generation through Static Analysis and Fuzz TestingThe First Workshop on Safe & Trustworthy Agents (SATA) at NeurIPS, 2024
- On the Design of Novel Attention Mechanism for Enhanced Efficiency of Transformers61st ACM Design Automation Conference (DAC), 2024
- Exploring the Predictive Capabilities of AlphaFold Using Adversarial Protein SequencesIEEE Transactions on Artificial Intelligence (IEEE TAI), 2024
- Counterexample Guided Inductive Synthesis Using Large Language Models and Satisfiability SolvingIEEE Military Communications Conference (MILCOM), 2023BEST PAPER CANDIDATE
- Automated Synthesis for In-Memory Computing42nd International Conference on Computer-Aided Design (ICCAD), 2023
- UpTime: Towards Flow-Based In-Memory Computing with High Fault-Tolerance60th Design Automation Conference (DAC), 2023
- Logic Synthesis for Digital In-Memory Computing41st International Conference on Computer-Aided Design (ICCAD), 2022BEST PAPER CANDIDATE
- Shaping Noise for Robust Attributions in Neural Stochastic Differential Equations36th AAAI International Conference on Artificial Intelligence (AAAI), 2022Oral Presentation
- PATH: Evaluation of Boolean Logic using Path-based In-Memory Computing59th Design Automation Conference (DAC), 2022
- Towards Resilient Analog In-Memory Deep Learning via Data Layout Re-Organization59th Design Automation Conference (DAC), 2022
- On Smoother Attributions using Neural Stochastic Differential Equations30th International Joint Conference on Artificial Intelligence (IJCAI), 2021
- Hybrid Analog-Digital In-Memory ComputingInternational Conference on Computer-Aided Design (ICCAD), 2021
- COMPACT: Flow-Based Computing on Nanoscale Crossbars with Minimal SemiperimeterDesign Automation and Test in Europe (DATE), 2021BEST PAPER CANDIDATE
- DP-MAP: Towards Resistive Dot-Product Engines with Improved PrecisionInternational Conference on Computer-Aided Design (ICCAD), 2020
- Attribution-Based Confidence Metric for Deep Neural NetworksAdvances in Neural Information Processing Systems (NeurIPS), 2019