Susmit Jha

EDUCATION

University of California, Berkeley — Berkeley, CA

Ph.D., Computer Science — August 2006 - December 2011

• Partially supported by Berkeley Fellowship
• GPA: 4.0/4.0
• Thesis: Towards Automated System Synthesis Using Structurally Constrained Induction and Deduction
• Leon O. Chua Award
• 10-Year Most Influential Paper Award, 42nd International Conference on Software Engineering (ICSE), 2020

M.S., Computer Science — August 2006 - December 2011

• GPA: 4.0/4.0
• Thesis: Reachability Analysis of Lazy Linear Hybrid Automata

Indian Institute of Technology, Kharagpur — Kharagpur, India

B.Tech., Computer Science and Engineering — August 2002 - May 2006

• GPA: 9.74/10.0
• TCS Gold Medal

PROFESSIONAL EXPERIENCE

SRI International (Stanford Research Institute) — Menlo Park, CA

Technical Director — 2022 - 2025

Principal Scientist — 2019 - 2022

Senior Scientist — 2016 - 2019

• Established and led the Neuro-symbolic Computing and Intelligence Research Group
• Principal Investigator for DARPA, IARPA, ARL, NSA, and NSF-funded projects totaling over $26M.
• Crucial contributor for additional $29M funding.
• Service on the DARPA ISAT Study Group 2023-2026.
• Co-inventor of NEDL in DARPA SoSITE that led to STITCHES Air Force program of record.
• Led projects with universities as subcontractors, including MIT, Caltech, CMU, UT Austin, Boston University, Stony Brook University, and the University of Houston.
• Hired 8 full-time AI researchers in the Computer Science Laboratory at SRI
• Hosted more than 20 student interns in the research group, with many co-authoring research publications; 3 interns became full-time researchers in the group.
• Nominated for SRI Achievement Awards: Change the World – unclassified (2019), Living our Mission and Values (Manager/Leader) (2022), Living our Mission and Values (individual) (2023, 2024).

DARPA ISAT Member — 2023-2025

• Led TRACE ISAT Study (2023-2024) on Large Language Models for Verifiable Artifacts.
• Led STORM ISAT Study (2024-2025) on Stochastic Computing for Energy-Efficient AI.
• Led SCALEBREAKER ISAT Study (2024-2025) on new AI architectures that disrupt LLM Scaling Laws.

United Technologies Research Center (now RTRC, Raytheon) — Berkeley, CA, USA

Staff Research Scientist — December 2014 - October 2016

• PI for Explainable Artificial Intelligence IRAD, leading to novel methods for logic-based (predicate logic, temporal logic, deontic logic) interrogation of deep learning models, including RL policies.
• PI for Automated Synthesis of Switching Logic for Cyberphysical Systems IRAD project.
• Automated verification and synthesis of code for software-defined radio in DARPA Communication in Contested Environments; received UTRC Great Job Award.
• Developed assume-guarantee style human-machine interaction models in DARPA Aircrew Labor In-Cockpit Automation System.

Intel Corporation — Hillsboro, OR, USA

Research Scientist — January 2012 - December 2014

• Game-theoretic and multi-arm bandit optimization for online customization of platform-level user experience and performance.
• Received Research Technology Scoping Award in 2014 for defining a new research space of platform-level debugging of energy and performance bugs as an out-of-distribution detection challenge.
• Received Division Recognition Award in 2012 for contributions to predictive models for Systems on Chip (SoC), including network-on-chip performance modeling.
• Authored multiple papers in Intel’s Design and Test Technology Conference on combining formal models with machine learning to model networks on chip.

RESEARCH OVERVIEW AND FUNDING SUMMARY

Before joining DARPA, Dr. Susmit Jha was a Technical Director in the Computer Science Laboratory at SRI International, where he led the research group on Neuro-symbolic Computing and Intelligence with research funded by DARPA, IARPA, ARPA-H, NSA, ARL, DLA, and NSF. His research focuses on trustworthy AI and formal methods for high-assurance applications such as autonomy, cyber-security, and AI-assisted design and engineering.

• Over 100 peer-reviewed publications.
• In the last 5 years, over 15 papers in top CSRankings venues such as NeurIPS, ICLR, ICML, AAAI, DAC, ICSE, CVPR, ICCV, TAI, TMLR, and JMLR
• Over 5300 citations; h-index 35; i10-index 74.
• Secured over $26M as Principal Investigator and contributed to an additional $29M in funding.

Selected Projects Led as Principal Investigator

• DARPA Assured Neuro-Symbolic Learning and Reasoning (ANSR)
  Role: Principal Investigator
  Contract #: FA8750-23-C-0519
  Team: SRI (Prime), USMA West Point, CMU, UCLA
  Contract Value: $5.39M + $1M for USMA
• DARPA Symbiotic Design on NeuroSymbolic Design of Cyberphysical Systems (SDCPS)
  Role: Principal Investigator
  Contract #: FA8750-20-C-0002
  Team: SRI (Prime), UT Austin, Penn State University, CMU
  Contract Value: $5.62M
• DARPA Transfer from Imprecise and Abstract Models to Autonomous Technologies (TIAMAT)
  Role: Principal Investigator
  Contract #: HR0011-24-9-0424
  Team: SRI (Prime)
  Contract Value: $2.07M
• DARPA Assured Autonomy (AA)
  Role: Principal Investigator
  Contract #: FA8750-19-C-0089
  Team: SRI (Prime), Caltech, MIT
  Contract Value: $2M
• IARPA Trojans in AI (TrojAI)
  Role: Principal Investigator
  Contract #: W911NF-20-C-0038
  Team: SRI (Prime), Boston University, Stony Brook University, University of Houston, UC Davis
  Contract Value: $6.81M
• ARL Alliance for IoBT Research on Evolving Intelligent Goal-Driven Networks (IoBT REIGN)
  Role: Institute Principal Investigator
  Contract #: W911NF-17-2-0196
  Team: UIUC (Prime); SRI sub to UIUC
  Contract Value for SRI: $1.85M [ $1.4M (2018-2023) + $451K (2023-2026) ]
• NSF CPS: Small: Self-Improving Cyber-Physical Systems
  Role: Principal Investigator
  NSF Award #: 1740079
  Team: SRI
  Grant Value: $500K
• NSF EAGER: Duality-Based Algorithm Synthesis
  Role: Co-Principal Investigator
  NSF Award #: 1750009
  Team: SRI
  Grant Value: $250K
• NSA-GTRI Trinity4Cyber: Safe Generative AI for Cyber-security
  Role: Institute Principal Investigator
  Team: GTRI (Prime)
  Contract Value: $1.55M

Total Funding as Principal Investigator: $26.04M

Dr. Jha’s research led to the co-development of NEDL technology in DARPA SoSITE, which became the foundation for the STITCHES Air Force Program of Record.

He was named to the DARPA ISAT Study Group for 2023-2026 and left the group in 2025 when he became a Program Manager at DARPA.

HONORS, RECOGNITIONS, AND AWARDS

Major Awards and Recognitions

• 10-Year Most Influential Paper Award, IEEE/ACM 42nd International Conference on Software Engineering (ICSE), 2020
• Top 10% Paper Award, 17th International Design Conference (DESIGN), 2022
• Best Paper Award Nomination, IEEE/ACM 15th International Conference on Cyber-Physical Systems (ICCPS), 2025
• Best Paper Award Nomination, IEEE 4th International Conference on AI Engineering – Software Engineering for AI, 2025
• Best Paper Award Nomination, IEEE/ACM 14th International Conference on Cyber-Physical Systems (ICCPS), 2023
• Best Paper Award Nomination, IEEE 41st Military Communications Conference (MILCOM), 2023
• Best Demonstration Award for Sherlock - A Tool For Verification Of Neural Network Feedback Systems at 22nd ACM International Conference on Hybrid Systems: Computation and Control (HSCC), 2019
• SRI Achievement Awards Nominations [2019, 2022, 2023, 2024]
• UTRC Great Job Award, United Technologies Research Center, Berkeley, 2016
• Research Technology Scoping Award, Intel Corporation, 2014
• Best Speaker Award, International Workshop on Numerical Software Verification, 2013
• Division Recognition Award, Intel Corporation, 2012
• Leon O. Chua Award, University of California, Berkeley, 2011
• Travel Awards to PLDI 2011 and CAV 2009
• Berkeley Fellowship for Graduate Studies, University of California, Berkeley, 2006-2008
• TCS Gold Medal, Indian Institute of Technology (IIT), Kharagpur, 2006

INVITED TALKS, KEYNOTE LECTURES AND TUTORIALS

• Invited Talk at CodeBot’25 – Can We Trust AI-Generated Code? Workshop organized by US Army Research Laboratory, 2025
• Invited Talk on Neuro-Symbolic Generative AI Assistant for System Design at 22nd ACM-IEEE International Symposium on Formal Methods and Models for System Design (MEMOCODE), 2024
• Keynote Talk at NSF Workshop on Hardware-Software Co-design for Neuro-Symbolic Computation, 2024
• Keynote Talk on TrinityAI: towards trustworthy, resilient, and interpretable AI for high-assurance applications at SPIE Assurance and Security for AI-enabled Systems, 2024
• Keynote Talk on Neuro-symbolic Learning with Large Foundation Models at INFORMS Business Analytics Conference, 2023
• Keynote Talk and Tutorial at Design Summer School at University of Maryland, College Park, 2023
• Tutorials at multiple iterations of NSF Summer School in Formal Methods organized at SRI.
• Invited panels at ACM/IEEE Symposium on Edge Computing (2023) and Design Automation Conference (2020).
• Tutorials at DAC 2023, C3E 2023, Post-Industrial Summit on AI Agents 2024, IDETC-CIE 2022, and others.
• Invited Panel at NSF Convergence Accelerator Workshop: Provably Safe and Beneficial Artificial Intelligence (PSBAI), 2022
• Invited Talk on Trinity: On Trust, Resilience and Interpretability of AI at Stanford Center for AI Safety, 2021
• Invited Talk at ISAT/DARPA DRACULA Study “Designing and Resisting AI Cyber User-in-the-Loop Attacks”, 2021
• Invited Talk on Formal Synthesis: Oracle-guided Learning of Compositional Concepts at The Simons Institute for the Theory of Computing, 2021
• Tutorials at workshops organized by Thrive-WISE 2019-2024

SELECTED CONFERENCE PUBLICATIONS WITH ACCEPTANCE RATES

• [S25] Malyaban Bal, Brian Matejek, Susmit Jha, and Adam Cobb. SpikingVTG: A Spiking Detection Transformer for Video Temporal Grounding. NeurIPS 2025. Acceptance Rate: 24.52%.
• [C24] Ayush Gupta, Anirban Roy, Rama Chellappa, Nathaniel D. Bastian, Alvaro Velasquez, and Susmit Jha. TOGA: Temporally Grounded Open-Ended Video QA with Weak Supervision. ICCV 2025. Acceptance Rate: 24%.
• [C23] Ayush Gupta, Ramneet Kaur, Anirban Roy, Adam D. Cobb, Rama Chellappa, and Susmit Jha. Polysemantic Dropout: Conformal OOD Detection for Specialized LLMs. EMNLP 2025. Acceptance Rate: 22.16%.
• [S22] C. Spiess et al. Calibration and Correctness of Language Models for Code. ICSE 2025. Acceptance Rate: 22.3%.
• [S21] A. D. Cobb et al. Direct Amortized Likelihood Ratio Estimation. AAAI 2024. Acceptance Rate: 23.75%.
• [S20] W. Lyu et al. Task-agnostic detector for insertion-based backdoor attacks. NAACL Findings 2024. Acceptance Rate: 12.5%.
• [S19] R. Kaur et al. CODiT: Conformal Out-of-Distribution Detection in Time-Series Data for Cyber-Physical Systems. ICCPS 2023. Acceptance Rate: 25%.
• [S18] A. D. Cobb et al. AircraftVerse: A Large-Scale Multimodal Dataset of Aerial Vehicle Designs. NeurIPS 2023. Acceptance Rate: 26.5%.
• [S17] S. K. Jha et al. Shaping Noise for Robust Attributions in Neural Stochastic Differential Equations. AAAI 2022. Acceptance Rate: 15%.
• [S16] R. Kaur et al. iDECODe: In-Distribution Equivariance for Conformal Out-of-Distribution Detection. AAAI 2022. Acceptance Rate: 15%.
• [S14] X. Hu et al. Trigger Hunting with a Topological Prior for Trojan Detection. ICLR 2022. Acceptance Rate: 32.9%.
• [S13] E. Cunningham, A. D. Cobb, and S. Jha. Principal Component Flows. ICML 2022. Acceptance Rate: 21.9%.
• [S12] M. Acharya et al. Detecting Out-of-Context Objects Using Graph Contextual Reasoning Network. IJCAI 2022. Acceptance Rate: 14.9%.
• [S11] S. K. Jha et al. ExplainIt!: A Tool for Computing Robust Attributions of DNNs. IJCAI 2022. Acceptance Rate: 14.9%.
• [S10] S. K. Jha et al. On Smoother Attributions Using Neural Stochastic Differential Equations. IJCAI 2021. Acceptance Rate: 13.9%.
• [S9] T. Sahai et al. Estimating the Density of States of Boolean Satisfiability Problems on Classical and Quantum Computing Platforms. AAAI 2020. Acceptance Rate: 20.6%.
• [S8] P. Kiourti et al. TrojDRL: Evaluation of Backdoor Attacks on Deep Reinforcement Learning. DAC 2020. Acceptance Rate: 23.1%.
• [S7] U. Jang, S. Jha, and S. Jha. On the Need for Topology-Aware Generative Models for Manifold-Based Defenses. ICLR 2020. Acceptance Rate: 26.5%.
• [S6] S. Jha et al. Attribution-Based Confidence Metric for Deep Neural Networks. NeurIPS 2019. Acceptance Rate: 21.1%.
• [S5] M. Vazquez-Chanlatte et al. Learning Task Specifications from Demonstrations. NeurIPS 2018. Acceptance Rate: 20.8%.
• [S4] S. Gulwani, S. Jha, A. Tiwari, and R. Venkatesan. Synthesis of Loop-Free Programs. PLDI 2011. Acceptance Rate: 23.3%.
• [S3] S. Jha et al. Oracle-Guided Component-Based Program Synthesis. ICSE 2010. Acceptance Rate: 14%.
• [S2] S. Jha, R. Limaye, and S. A. Seshia. Beaver: Engineering an Efficient SMT Solver for Bit-Vector Arithmetic. CAV 2009. Acceptance Rate: 31%.
• [S1] C. Sturton et al. On voting machine design for verification and testability. CCS 2009. Acceptance Rate: 18%.

JOURNAL PUBLICATIONS

• [J15] Brian Matejek et al. SAFE-NID: Self-Attention with Normalizing-Flow Encodings for Network Intrusion Detection. Transactions on Machine Learning Research, 2025.
• [J14] Alexander M. Berenbeim et al. Applications of Certainty Scoring for Machine Learning Classification and Out-of-Distribution Detection. ACM Transactions on Probabilistic Machine Learning, 2025.
• [J13] I. R. Alkhouri et al. Exploring the Predictive Capabilities of AlphaFold Using Adversarial Protein Sequences. IEEE Trans. Artif. Intell., 2024.
• [J12] A. Magesh et al. Principled Out-of-Distribution Detection via Multiple Testing. Journal of Machine Learning Research, 2023.
• [J11] A. Cobb et al. On Diverse System-Level Design Using Manifold Learning and Partial Simulated Annealing. Proceedings of the Design Society, 2022.
• [J10] S. Jha et al. TeLEx: Learning Signal Temporal Logic from Positive Examples Using Tightness Metric. Formal Methods Syst. Des., 2019.
• [J9] S. Dutta et al. Learning and Verification of Feedback Control Systems using Feedforward Neural Networks. IFAC-PapersOnLine, 2018.
• [J8] S. Jha et al. Explaining AI Decisions Using Efficient Methods for Learning Sparse Boolean Formulae. J. Autom. Reason., 2019.
• [J7] T. F. Abdelzaher et al. Toward an Internet of Battlefield Things: A Resilience Perspective. Computer, 2018.
• [J6] S. Jha et al. Safe Autonomy Under Perception Uncertainty Using Chance-Constrained Temporal Logic. J. Autom. Reason., 2018.
• [J5] S. Jha and S. A. Seshia. A Theory of Formal Synthesis via Inductive Learning. Acta Informatica, 2017.
• [J4] F. Hussain et al. Parameter Discovery in Stochastic Biological Models Using Simulated Annealing and Statistical Model Checking. Int. J. Bioinform. Res. Appl., 2014.
• [J3] A. K. Ghosh et al. Discovering Rare Behaviours in Stochastic Differential Equations Using Decision Procedures: Applications to a Minimal Cell Cycle Model. Int. J. Bioinform. Res. Appl., 2014.
• [J2] S. K. Jha et al. Synthesis of Insulin Pump Controllers from Safety Specifications Using Bayesian Model Validation. Int. J. Bioinform. Res. Appl., 2012.
• [J1] S. Jha and R. K. Shyamasundar. Adapting Biochemical Kripke Structures for Distributed Model Checking. Trans. Comp. Sys. Biology, 2006.

EDITED VOLUMES AND BOOK CHAPTERS

• [B7] Message from DESTION 2024 Chairs. 2024 IEEE Workshop on Design Automation for CPS and IoT (DESTION), 2024.
• [B6] N. D. Bastian, S. Jha, P. Tabuada, V. Veeravalli, and G. Verma. Principles of Robust Learning and Inference for IoBTs. In: IoT for Defense and National Security, 2023.
• [B5] DESTION 2022 Committee. 2022 IEEE Workshop on Design Automation for CPS and IoT (DESTION), 2022.
• [B4] S. Jha et al. Trinity: Trust, Resilience and Interpretability of Machine Learning Models. In: Game Theory and Machine Learning for Cyber Security, 2021.
• [B3] R. Lee, S. Jha, and A. Mavridou (Eds.). NASA Formal Methods - 12th International Symposium (NFM 2020). LNCS 12229, Springer, 2020.
• [B2] S. A. Seshia, X. Zhu, A. Krause, and S. Jha. Machine Learning and Formal Methods (Dagstuhl Seminar 17351). Dagstuhl Reports, 2018.
• [B1] K. Chatterjee, R. Ehlers, and S. Jha (Eds.). Proceedings 3rd Workshop on Synthesis (SYNT 2014). EPTCS 157, 2014.

CONFERENCE PUBLICATIONS

• [C82] Bhusal, Bishnu, Manoj Acharya, Ramneet Kaur, Colin Samplawski, Anirban Roy, Adam D. Cobb, Rohit Chadha, and Susmit Jha. ”Privacy Preserving In-Context-Learning Framework for Large Language Models.” In Association for the Advancement of Artificial Intelligence (AAAI), 2026
• [C81] Velasquez, Alvaro, Susmit Jha, and Ismail R. Alkhouri. ”On the Dataless Training of Neural Networks.” Association for the Advancement of Artificial Intelligence - Emerging Trends in AI (AAAI-ETA), 2026
• [C80] Malyaban Bal, Brian Matejek, Susmit Jha, and Adam Cobb. ”SpikingVTG: A Spiking Detection Transformer for Video Temporal Grounding.” In The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS), 2025
• [C79] Gupta, Ayush, Anirban Roy, Rama Chellappa, Nathaniel D. Bastian, Alvaro Velasquez, and Susmit Jha. ”TOGA: Temporally Grounded Open-Ended Video QA with Weak Supervision.” In International Conference on Computer Vision (ICCV), 2025.
• [C78] Gupta, Ayush, Ramneet Kaur, Anirban Roy, Adam D. Cobb, Rama Chellappa, and Susmit Jha. ”Polysemantic Dropout: Conformal OOD Detection for Specialized LLMs.” In Conference on Empirical Methods in Natural Language Processing (EMNLP), 2025
• [C77] Lin, Vivian, Ramneet Kaur, Yahan Yang, Souradeep Dutta, Yiannis Kantaros, Anirban Roy, Susmit Jha, Oleg Sokolsky, and Insup Lee. ”Safety monitoring for learning-enabled cyber-physical systems in out-of- distribution scenarios.” In Proceedings of the ACM/IEEE 16th International Conference on Cyber-Physical Systems (with CPS-IoT Week 2025)(ICCPS), pp. 1-11. 2025.
• [C76] Baratta, Laura R., Sunny S. Lou, Thomas Kannampallil, Susmit Jha, Anirban Roy, and Adam D. Cobb. ”Predicting Secure Messaging Traffic in Clinical Settings.” Studies in health technology and informatics 329: 553-557, 2025.
• [C75] C. Samplawski, A. Cobb, M. Acharya, R. Kaur and S. Jha. ”Scalable Bayesian Low-Rank Adaptation of Large Language Models via Stochastic Variational Subspace Inference” in Uncertainty in AI (UAI), 2025
• [C74] C Nagesh. S. Sankaranarayanan, R. Kaur, T. Sahai and S. Jha. ”Taylor-Model Physics-Informed Neural Networks (PINNs) for Ordinary Differential Equations” in International Conference on Neuro-symbolic Systems (NeuS), 2025
• [C73] C. Spiess, D. Gros, K. S. Pai, M. Pradel, M. R. I. Rabin, A. Alipour, S. Jha, P. Devanbu, and T. Ahmed, ”Calibration and Correctness of Language Models for Code,” in International Conference on Software Engineering (ICSE), 2025.
• [C72] B Hu, D Gopinath, R Mangal, N Narodytska, C Pasareanu, S, Jha, ”Debugging and Runtime Analysis of Neural Networks with VLMs,” in 4th International Conference on AI Engineering – Software Engineering for AI (CAIN), 2025.
• [C71] F. Toledo, S. Elbaum, D. Gopinath, R. Kaur, R. Mangal, C. S. Pa˘sa˘reanu, A. Roy, S. Jha, ”Monitoring Safety Properties for Autonomous Driving Systems with Vision-Language Models ,” in IEEE International Conference on Engineering Reliable Autonomous Systems (ERAS), 2025.
• [C70] A. D. Cobb, B. Matejek, D. Elenius, A. Roy, and S. Jha, ”Direct Amortized Likelihood Ratio Estimation,” in Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), vol. 38, no. 18, pp. 20362-20369, 2024. doi:10.1609/AAAI.V38I18.30018
• [C69] W. Lyu, X. Lin, S. Zheng, L. Pang, H. Ling, S. Jha, and C. Chen, ”Task-Agnostic Detector for Insertion- Based Backdoor Attacks,” in Proceedings of the North American Chapter of the Association for Computational Linguistics - Human Language Technologies (NAACL-HLT), Mexico City, Mexico, pp. 2808-2822, 2024. doi:10.18653/v1/2024.findings-naacl.179
• [C68] S. K. Jha, S. Jha, R. Ewetz, and A. Velasquez, ”On the Design of Novel Attention Mechanism for Enhanced Efficiency of Transformers,” in 61st ACM Design Automation Conference (DAC), 2024.
• [C67] R. Mangal, N. Narodytska, D. Gopinath, B. C. Hu, A. Roy, S. Jha, and C. S. Pa˘ sa˘ reanu, ”Concept-Based Analysis of Neural Networks via Vision-Language Models,” in Proceedings of the International Symposium on AI Verification (SAIV), Montreal, Canada, pp. 49-77, 2024. doi:10.1007/978-3-031-65112-0 3
• [C66] S. K. Jha, S. Jha, R. Ewetz, and A. Velasquez, ”Solving Mystery Planning Problems Using Category Theory, Functors, and Large Language Models,” in 3rd International Conference on Assured Autonomy (ICAA), 2024.
• [C65] S. K. Jha, S. Jha, M. R. H. Rashed, R. Ewetz, and A. Velasquez, ”Automated Synthesis of Hardware Designs using Symbolic Feedback and Grammar-Constrained Decoding in Large Language Models,” in IEEE National Aerospace and Electronics Conference (NAECON), pp. 95-100, 2024.
• [C64] S. Jha, ”TrinityAI: Towards Trustworthy, Resilient, and Interpretable AI for High-Assurance Applications,” in Assurance and Security for AI-enabled Systems, vol. 13054, pp. 1305402, 2024, SPIE.
• [C63] A. Magesh, V. Veeravalli, A. Roy, and S. Jha, ”Detection of Untrustworthy Outputs from Learning Models: A Multiple Testing Approach,” in Proceedings of the Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, CA, Oct. 27-30, 2024.
• [C62] A. M. Berenbeim, A. V. Wei, A. D. Cobb, A. Roy, S. Jha, and N. D. Bastian, ”Bayesian Graph Representation Learning for Adversarial Patch Detection,” in Assurance and Security for AI-enabled Systems, vol. 13054, pp. 54-70, 2024, SPIE.
• [C61] S. Jha, ”Lightning Talk: Trinity - Assured Neuro-symbolic Model Inspired by Hierarchical Predictive Coding,” in Proceedings of the Design Automation Conference (DAC), San Francisco, CA, USA, pp. 1-2, 2023. doi:10.1109/DAC56929.2023.10247803
• [C60] R. Kaur, S. Jha, A. Roy, O. Sokolsky, and I. Lee, ”Predicting Out-of-Distribution Performance of Deep Neural Networks Using Model Conformance,” in Proceedings of the IEEE International Conference on Assured Autonomy (ICAA), Laurel, MD, USA, pp. 19-28, 2023. doi:10.1109/ICAA58325.2023.00011
• [C59] K. Sikka, I. Sur, A. Roy, A. Divakaran, and S. Jha, ”Detecting Trojaned DNNs Using Counterfactual Attributions,” in Proceedings of the IEEE International Conference on Assured Autonomy (ICAA), Laurel, MD, USA, pp. 76-85, 2023. doi:10.1109/ICAA58325.2023.00019
• [C58] S. Jha, S. K. Jha, P. Lincoln, N. D. Bastian, A. Velasquez, and S. Neema, ”Dehallucinating Large Language Models Using Formal Methods Guided Iterative Prompting,” in Proceedings of the IEEE International Conference on Assured Autonomy (ICAA), Laurel, MD, USA, pp. 149-152, 2023. doi:10.1109/ICAA58325.2023.00029
• [C57] R. Kaur, K. Sridhar, S. Park, Y. Yang, S. Jha, A. Roy, O. Sokolsky, and I. Lee, ”CODiT: Conformal Out-of-Distribution Detection in Time-Series Data for Cyber-Physical Systems,” in Proceedings of the ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS), San Antonio, TX, USA, pp. 120-131, 2023. doi:10.1145/3576841.3585931
• [C56] I. Sur, K. Sikka, M. Walmer, K. Koneripalli, A. Roy, X. Lin, A. Divakaran, and S. Jha, ”TIJO: Trigger Inversion with Joint Optimization for Defending Multimodal Backdoored Models,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Paris, France, pp. 165-175, 2023. doi:10.1109/ICCV53048.2023.00022
• [C55] A. Magesh, V. V. Veeravalli, A. Roy, and S. Jha, ”Principled OOD Detection via Multiple Testing,” in Proceedings of the IEEE International Symposium on Information Theory (ISIT), Taipei, Taiwan, pp. 1026-1031, 2023. doi:10.1109/ISIT54713.2023.10206581
• [C54] S. Jha, A. Roy, A. D. Cobb, A. M. Berenbeim, and N. D. Bastian, ”Challenges and Opportunities in Neuro- Symbolic Composition of Foundation Models,” in Proceedings of the IEEE Military Communications Conference (MILCOM), Boston, MA, USA, pp. 156-161, 2023. doi:10.1109/MILCOM58377.2023.10356344
• [C53] S. K. Jha, S. Jha, R. Ewetz, and A. Velasquez, ”Neural SDEs for Robust and Explainable Analysis of Electromagnetic Unintended Radiated Emissions,” in Proceedings of the IEEE Military Communications Conference (MILCOM), Boston, MA, USA, pp. 655-660, 2023. doi:10.1109/MILCOM58377.2023.10356358
• [C52] S. K. Jha, S. Jha, P. Lincoln, N. D. Bastian, A. Velasquez, R. Ewetz, and S. Neema, ”Counterexample Guided Inductive Synthesis Using Large Language Models and Satisfiability Solving,” in Proceedings of the IEEE Military Communications Conference (MILCOM), Boston, MA, USA, pp. 944-949, 2023. doi:10.1109/MILCOM58377.2023.10356332
• [C51] A. B. Cobb, A. Roy, D. Elenius, F. M. Heim, B. Swenson, S. Whittington, J. D. Walker, T. Bapty, J. Hite, K. Ramani, C. McComb, and S. Jha, ”AircraftVerse: A Large-Scale Multimodal Dataset of Aerial Vehicle Designs,” in Proceedings of the Conference on Neural Information Processing Systems (NeurIPS), New Orleans, LA, USA, 2023.
• [C50] R. Kaur, S. Jha, A. Roy, S. Park, E. Dobriban, O. Sokolsky, and I. Lee, ”iDECODe: In-Distribution Equivariance for Conformal Out-of-Distribution Detection,” in Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), vol. 36, no. 7, pp. 7104-7114, 2022. doi:10.1609/aaai.v36i7.20670
• [C49] S. K. Jha, R. Ewetz, A. Velasquez, A. Ramanathan, and S. Jha, ”Shaping Noise for Robust Attributions in Neural Stochastic Differential Equations,” in Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), vol. 36, no. 9, pp. 9567-9574, 2022. doi:10.1609/aaai.v36i9.21190
• [C48] M. Walmer, K. Sikka, I. Sur, A. Shrivastava, and S. Jha, ”Dual-Key Multimodal Backdoors for Visual Question Answering,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA, pp. 15354-15364, 2022. doi:10.1109/CVPR52688.2022.01494
• [C47] X. Hu, X. Lin, M. Cogswell, Y. Yao, S. Jha, and C. Chen, ”Trigger Hunting with a Topological Prior for Trojan Detection,” in Proceedings of the International Conference on Learning Representations (ICLR), Virtual Event, 2022.
• [C46] E. Cunningham, A. D. Cobb, and S. Jha, ”Principal Component Flows,” in Proceedings of the International Conference on Machine Learning (ICML), Baltimore, MD, USA, pp. 4492-4519, 2022.
• [C45] M. Acharya, A. Roy, K. Koneripalli, S. Jha, C. Kanan, and A. Divakaran, ”Detecting Out-Of-Context Objects Using Graph Contextual Reasoning Network,” in Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), Vienna, Austria, pp. 629-635, 2022. doi:10.24963/ijcai.2022/89
• [C44] S. K. Jha, A. Velasquez, R. Ewetz, L. Pullum, and S. Jha, ”ExplainIt!: A Tool for Computing Robust Attributions of DNNs,” in Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), Vienna, Austria, pp. 5916-5919, 2022. doi:10.24963/ijcai.2022/853
• [C43] T. F. Abdelzaher, N. D. Bastian, S. Jha, L. M. Kaplan, M. B. Srivastava, and V. V. Veeravalli, ”Context- aware Collaborative Neuro-Symbolic Inference in IoBTs,” in Proceedings of the IEEE Military Communications Conference (MILCOM), Rockville, MD, USA, pp. 1053-1058, 2022. doi:10.1109/MILCOM55135.2022.10017607
• [C42] A. Roy, A. D. Cobb, N. D. Bastian, B. Jalaian, and S. Jha, ”Runtime Monitoring of Deep Neural Networks Using Top-Down Context Models Inspired by Predictive Processing and Dual Process Theory,” in AAAI Spring Symposium 2022, April 2022.
• [C41] P. Kiourti, W. Li, A. Roy, K. Sikka, and S. Jha, ”MISA: Online Defense of Trojaned Models using Misattributions,” in Proceedings of the Annual Computer Security Applications Conference (ACSAC), Virtual Event, pp. 570-585, 2021.
• [C40] S. Jha and A. Roy, ”On Detection of Out of Distribution Inputs in Deep Neural Networks,” in Proceedings of the IEEE International Conference on Cognitive Machine Intelligence (CogMI), Virtual Event, pp. 282-288, 2021.
• [C39] S. K. Jha, R. Ewetz, A. Velasquez, and S. Jha, ”On Smoother Attributions using Neural Stochastic Differential Equations,” in Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), Montreal, QC, Canada, pp. 522-528, 2021. doi:10.24963/ijcai.2021/72
• [C38] A. Velasquez, S. K. Jha, R. Ewetz, and S. Jha, ”Automated Synthesis of Quantum Circuits Using Symbolic Abstractions and Decision Procedures,” in Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS), Daegu, Korea (South), pp. 1-5, 2021. doi:10.1109/ISCAS51556.2021.9401250
• [C37] T. Sahai, A. Mishra, J. M. Pasini, and S. Jha, ”Estimating the Density of States of Boolean Satisfiability Problems on Classical and Quantum Computing Platforms,” in Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), New York, NY, USA, pp. 1627-1635, 2020. doi:10.1609/aaai.v34i02.5524
• [C36] D. Sun, S. Jha, and C. Fan, ”Learning Certified Control Using Contraction Metric,” in Proceedings of the Conference on Robot Learning (CoRL), Cambridge, MA, USA, pp. 1519-1539, 2020.
• [C35] P. Kiourti, K. Wardega, S. Jha, and W. Li, ”TrojDRL: Evaluation of Backdoor Attacks on Deep Reinforcement Learning,” in Proceedings of the ACM/IEEE Des

SELECTED PEER-REVIEWED WORKSHOP PAPERS

• [W18] X. Lin, M. Acharya, A. Roy, S. Jha, ”TeleLoRA: Teleporting Alignment across Large Language Models for Trojan Mitigation,” ICLR Workshop on Weight Space Learning, 2025
• [W17] L. R. Baratta, S. S. Lou, T. Kannampallil, S. Jha, A. Roy, A. Cobb , ”Predicting Secure Messaging Traffic in Clinical Settings,” MEDINFO, 2025
• [W16] A. Cobb, A. G. Baydin, B. A. Pearlmutter, S. Jha, ”Second-Order Forward-Mode Automatic Differentiation for Optimization,” NeurIPS 16th International OPT Workshop on Optimization for Machine Learning, 2024.
• [W15] M. Bal, B. Matejek, S. Jha, A. Cobb, ”SpikingVTG: Saliency Feedback Gating Enabled Spiking Video Temporal Grounding,” NeurIPS Machine Learning and Compression Workshop, 2024.
• [W14] R. Kaur, C. Samplawski, A. D. Cobb, A. Roy, B. Matejek, M. Acharya, D. Elenius, A. Michael Berenbeim, J. A. Pavlik, N. D. Bastian, S. Jha, ”Enhancing Semantic Clustering for Uncertainty Quantification & Conformal Prediction by LLMs,” NeurIPS Workshop on Statistical Frontiers in LLMs and Foundation Models @ NeurIPS 2024, 2024.
• [W13] R. Kaur, C. Samplawski, A. D. Cobb, A. Roy, B. Matejek, M. Acharya, D. Elenius, A. M. Berenbeim, J. A. Pavlik, N. D. Bastian, and S. Jha, ”Addressing Uncertainty in LLMs to Enhance Reliability in Generative AI,” NeurIPS Safe Generative AI Workshop, 2024.
• [W12] M. Acharya, X. Lin, and S. Jha, ”Investigating LLM Memorization: Bridging Trojan Detection and Training Data Extraction,” NeurIPS Safe Generative AI Workshop, 2024.
• [W11] M. Acharya, W. Zhou, A. Roy, X. Lin, W. Li, and S. Jha, ”Universal Trojan Signatures in Reinforcement Learning,” NeurIPS 2023 Workshop on Backdoors in Deep Learning-The Good, the Bad, and the Ugly.
• [W10] A. Cobb, A. Roy, D. Elenius, and S. Jha, ”Trinity AI Co-Designer for Hierarchical Oracle-Guided Design of Cyber-Physical Systems,” IEEE Workshop on Design Automation for CPS and IoT (DESTION), pp. 42–44, 2022.
• [W9] S. K. Jha, R. Ewetz, A. Velasquez, and S. Jha, ”Responsible Reasoning with Large Language Models and the Impact of Proper Nouns,” Workshop on Trustworthy and Socially Responsible Machine Learning, NeurIPS 2022.
• [W8] N. Le Van, A. Gehani, A. Gurfinkel, S. Jha, and J. A. Navas, ”Reinforcement Learning Guided Software Debloating,” in Workshop on ML for Systems at NeurIPS, 2019. [Online]. Available: mlforsystems.org
• [W7] S. Jha, ”Trust, Resilience, and Interpretability of AI Models,” Numerical Software Verification: 12th International Workshop, NSV 2019, New York City, NY, USA, July 13–14, 2019, Proceedings 12, pp. 3–25, 2019.
• [W6] D. Alrajeh, S. Jha, and S. Seshia, ”A Non-monotonic Theory of Oracle-guided Inductive Synthesis,” in LiVe 2017: 1st Workshop on Learning in Verification, Satellite Event of ETAPS 2017, Uppsala, Sweden, April 29, 2017.
• [W5] S. Jha, S. A. Seshia, and X. Zhu, ”On the Teaching Dimension of Octagons for Formal Synthesis,” 5th Workshop on Synthesis (SYNT), 2016.
• [W4] S. Jha and S. A. Seshia, ”Are There Good Mistakes? A Theoretical Analysis of CEGIS,” in Proceedings of the 3rd Workshop on Synthesis (SYNT), Vienna, Austria, pp. 84-99, 2014. doi:10.4204/EPTCS.157.10
• [W3] W. Li, S. Jha, and S. A. Seshia, ”Generating Control Logic for Optimized Soft Error Resilience,” 9th Workshop on Silicon Errors in Logic-System Effects (SELSE), 2013.
• [W2] S. Jha and S. A. Seshia, ”Synthesis of Optimal Fixed-Point Implementations of Numerical Software Routines,” in Numerical Software Verification (NSV), April 2013.
• [W1] S. Jha, W. Li, and S. A. Seshia, ”Localizing Transient Faults Using Dynamic Bayesian Networks,” 2009 IEEE International High Level Design Validation and Test (HLDVT) Workshop, pp. 82–87, 2009.

ACADEMIC MENTORSHIP OVERVIEW

• Mentor to 6 PhDs, 1 Master's, 1 MD in Neuro-symbolic Computing and Intelligence research groups.
• Mentored over 25 student interns, resulting in over 20 research papers with them.
• Served on Ph.D. thesis committees of Matthew Schuchhardt (Northwestern University) and Ramneet Kaur (University of Pennsylvania).
• Collaborator and mentor for non-profit Thrive-WISE, focused on retaining women in science and engineering; delivered multiple tutorials.
• Gave tutorials at multiple iterations of SRI Summer School in Formal Methods (NSF-funded) and Design Summer School at the University of Maryland, College Park.
• Presented tutorials at conferences and workshops such as DAC and NSV.

MENTORED STUDENTS

• Abhinav Verma, 2019
• Akshata Tiwari, 2024
• Aniket Roy, 2023
• Ayush Gupta, 2024
• Bishnu Bhusal, 2024
• Chitradeep Dutta Roy, 2019
• Hung Nguyen, 2025
• Laura Baratta, 2024
• Malyaban Bal, 2024
• Margaret P. Chapman, 2018
• Marcell J. Vazquez-Chanlatte, 2017
• Neelesh Verma, 2023
• Panagiota “Penny” Kiourti, 2019 and 2020
• Rohit Gupta, 2022
• SangHyuk Kim, 2024
• Seunghwan (Nigel) Kim, 2024
• Shromona Ghosh, 2018
• Soumitri Chattopadhyay, 2024
• Souradeep Dutta, 2017
• Stephen Giguere, 2019 and 2020
• Taha Belkhouja, 2022
• Trilok Padhi, 2024
• Uyeong Jang, 2018
• Weichao Zhou, 2019
• Matthew Schuchhardt, 2013 and 2014

PROFESSIONAL SERVICES

• DARPA ISAT Study Group Member (2023–2026). Serving on DARPA’s Information Science & Technology (ISAT) Study Group, which identifies emerging areas in computer science and information technology and recommends future research directions for the U.S. Department of Defense.
  • Lead, TRACE ISAT Study (2023–2024). Led the TRACE study on large language models for verifiable artifacts; briefed DARPA I2O leadership in summer 2024 and engaged participants from industry, government, and leading universities to identify opportunities for DARPA investment over the next decade.
  • Lead, STORM ISAT Study (2024–2025). Leading the STORM study on stochastic computing for energy-efficient AI; preliminary results briefed to DARPA I2O and DARPA MTO leadership in spring 2025, with final briefing planned for summer 2025.
  • Lead, SCALEBREAKER ISAT Study (2024–2025). Leading the SCALEBREAKER study on new AI architectures that disrupt LLM scaling laws; I2O program managers briefed in April 2025, with final briefing planned for summer 2025.
• Co-organizer, AAAI-24 Bridge Program “Artificial Intelligence for Design Problems” (2024).
• Invited Speaker, New Directions in Software Technology (NDIST) (2024).
• Co-organizer, AI Track, NSA-sponsored Computational Cybersecurity in Compromised Environments Workshop (2024).
• Speaker, Twenty-first High Confidence Software and Systems (HCSS) Conference (2024). Presented on Towards Creative Generative Models for Scientific Discovery.
• Co-organizer, Interdisciplinary Workshop on Trends in Human–AI Teaming for Engineering and Design, IDETC–CIE (ASME) (2023).
• Participant, Invite-only NSA-sponsored Computational Cybersecurity in Compromised Environments Workshop (2023).
• Workshop Program Co-chair, IEEE Workshop on Design Automation for CPS and IoT (DESTION) (2022). 2022 edition; IEEE, doi:10.1109/DESTION56136.2022.00006.
• Co-organizer, AI Track, NSA-sponsored Computational Cybersecurity in Compromised Environments Workshop (2022).
• Co-organizer, Interdisciplinary Workshop on Trends in Human–AI Teaming for Engineering and Design, IDETC–CIE (ASME) (2022).
• Workshop Program Co-chair, 2024 IEEE Workshop on Design Automation for CPS and IoT (DESTION) (2024). IEEE Computer Society, doi:10.1109/DESTION62938.2024.00005.
• Guest Co-editor, Special Issue on Artificial Intelligence and Cyber-Physical Systems, ACM Transactions on Cyber-Physical Systems (TCPS). Part 1: vol. 5, no. 4 (2021), DOI: 10.1145/3517045; Part 2: vol. 6, no. 2 (2022), DOI: 10.1145/3471164.
• Speaker, Twenty-first High Confidence Software and Systems (HCSS) Conference (2021). Presented on On Computing Relevant Parameters of Decision Functions.
• Program Co-chair, NASA Formal Methods Symposium (NFM 2020). Moffett Field, CA, USA, May 11–15, 2020.
• Guest Co-editor, Lecture Notes in Computer Science, vol. 12229 (2020). Springer, doi:10.1007/978-3-030-55754-6.
• Co-organizer, Dagstuhl Seminar “Machine Learning and Formal Methods” (Dagstuhl Seminar 17351) (2017). Report published in Dagstuhl Reports, vol. 7, issue 8, 2018, doi:10.4230/DagRep.7.8.i.
• Speaker, High Confidence Software and Systems (HCSS) Conference (2018). Presented on Data-driven Safe Control of Nonlinear Systems.
• Workshop Program Co-chair, 3rd Workshop on Synthesis (SYNT 2014). Vienna, Austria, July 23–24, 2014.
• Guest Co-editor, Electronic Proceedings in Theoretical Computer Science (EPTCS), vol. 157 (2014). doi:10.4204/EPTCS.157.
• Senior Area Chair / Area Chair / Meta-reviewer / PC Member / Reviewer (last six years). Service for leading conferences and workshops, including:
  • ICML 2022; CVPR 2022; AISTATS 2022; AAAI 2022; NeurIPS 2022
  • NASA Formal Methods 2023; FOMLAS 2023; AISTATS 2023; CVPR 2023; ICLR 2023
  • DESTION 2024; IJCAI 2024; AISTATS 2024; AAAI 2024; WACV 2024; NeurIPS 2024
  • NASA Formal Methods 2025; FMCAD 2025; NeuS 2025; WACV 2025; ICCV 2025; CVPR 2025; AAAI 2025; ICLR 2025; ICML 2025; NeurIPS 2025
  • AAAI 2026; ICLR 2026; ICML 2026; FASE 2026; AISTATS 2026; NFM 2026; WACV 2026; CVPR 2026
• Proposal Review Panelist. Served on multiple proposal review panels over the last decade for:
  • National Science Foundation (NSF)
  • Army Research Office (ARO)
  • US–Israel Binational Science Foundation (BSF)

PATENTS

• [P9] A data-driven surrogate model for predicting flow field properties around 3D objects
  Publication number: 20240312129 (Application)
  Filed: November 14, 2023 — Publication date: September 19, 2024
  Inventors: Anirban Roy, Adam Derek Cobb, Daniel Elenius, Patrick Denis Lincoln, Susmit Jha
• [P8] Diversity-aware multi-objective high dimensional parameter optimization using invertible models
  Publication number: 20240143689 (Application)
  Filed: October 18, 2023 — Publication date: May 2, 2024
  Inventors: Susmit Jha, Adam Derek Cobb, Anirban Roy, Daniel Elenius, Patrick Denis Lincoln
• [P7] Iterative bootstrapping neurosymbolic method for generating system designs
  Publication number: 20240169129 (Application)
  Filed: November 17, 2023 — Publication date: May 23, 2024
  Inventors: Adam Derek Cobb, Daniel Elenius, Anirban Roy, Patrick Denis Lincoln, Susmit Jha
• [P6] Multipath verification of data transforms in a system of systems
  Publication number: 20220197881 (Application)
  Filed: December 17, 2021 — Publication date: June 23, 2022
  Inventors: Bruno Dutertre, Susmit Jha, Huascar Sanchez, Patrick Lincoln, Eric M. Pearson, Richard Dean, Ian A. Mason
• [P5] Analysis and design of dynamical system controllers using neural differential equations
  Patent number: 12236330 (Grant)
  Filed: May 26, 2021 — Date of Patent: February 25, 2025
  Inventors: Ajay Divakaran, Anirban Roy, Susmit Jha
• [P4] Trusted neural network system
  Patent number: 11651227 (Grant)
  Filed: December 19, 2018 — Date of Patent: May 16, 2023
  Inventors: Shalini Ghosh, Patrick Lincoln, Ashish Tiwari, Susmit Jha
• [P3] Identifying fixed bits of a bitstring format
  Patent number: 11423247 (Grant)
  Filed: April 3, 2020 — Date of Patent: August 23, 2022
  Inventors: Ashish Tiwari, Susmit Jha, Patrick Lincoln
• [P2] Large Language Models for Quantum Transpiling
  Publication number: 20250036993 (Application)
  Filed: May 24, 2024 — Publication date: January 30, 2025
  Inventors: Adam Cobb, Susmit Jha
• [P1] Rare events estimation
  Publication number: 20250278650 (Application)
  Filed: December 18, 2024 — Publication date: September 4, 2025
  Inventors: Tuhin Sahai, Ahmed Aloui, Susmit Jha

COMMERCIALIZATION AND ADVISORY ROLES

• Technical Advisor, Confidencial, a startup focusing on cybersecurity of data for AI.
• Co-Founder and Senior Technical Advisor, P-1.ai, an AI startup working on artificial general engineering intelligence.
• Co-inventor of NEDL in DARPA SoSITE that led to STITCHES Air Force program of record.