Tergel Molom-Ochir

Tergel Molom-Ochir

Ph.D. Candidate
Electrical & Computer Engineering
Duke University

Research Associate, Hewlett Packard Labs — Emerging Accelerators Team

Email: tergel.molom-ochir@duke.edu

CV | Google Scholar | GitHub | LinkedIn | X

Building memristive associative memory hardware for the next generation of AI inference and reasoning systems.

Biography

I am a Ph.D. candidate in Electrical and Computer Engineering at Duke University, advised by Dr. Yiran Chen and Dr. Hai "Helen" Li in the Center for Computational Evolutionary Intelligence (CEI). I also serve as a Research Associate on the Emerging Accelerators team at Hewlett Packard Labs.

My research designs memristive associative memory hardware for AI inference and reasoning — spanning device cells, CAM circuits, and in-memory architectures for attention, search, and test-time compute. I collaborate closely with Prof. Leon Chua and Dr. R. Stanley Williams on volumetric memristor circuits, and with Dr. Qiangfei Xia at UMass Amherst on emerging memory devices.

Previously, I completed my B.S. in Electrical Engineering at UMass Amherst with Dr. Xia, and worked with Dr. Yingyan Lin at Rice University on efficient deep learning.

News

[May 2026] Started Research Associate position at HPE Labs (Emerging Accelerators Team).
[Apr 2026] I passed my Ph.D. preliminary exam at Duke ECE.
[Apr 2026] CAMformer accepted at IEEE TCAS-I.
[Mar 2026] DirectGeMM accepted at ISCAS 2026.
[Jan 2026] Began serving as manuscript reviewer for Nature Scientific Reports and Nature Computational Science.
[Dec 2025] NP-CAM accepted at HPCA 2025.
[Oct 2025] Patent filed at Hewlett Packard Enterprise for In-Memory MCTS Accelerator.
[May 2025] Started Research Associate position at HPE Labs (Emerging Accelerators Team).
[2025] CAM Survey published in IEEE TCAS-I; MonoSparse-CAM published at ISCAS 2025.
[Aug 2023] Joined Duke University CEI Lab.

Research Highlights

I design memristive associative memory hardware for AI inference and reasoning, working across the device–circuit–architecture stack to make associative computation a first-class primitive in modern AI systems.

Memristive Substrates

Device-level cells and volumetric arrays — 3D cellular neural networks for in-memory computing, RRAM characterization, in-memory ADC primitives.

with Chua, Williams, Xia

Associative Memory Circuits

CAM cells, arrays, and architectures — CAMformer, MonoSparse-CAM, NP-CAM, and a comprehensive survey of 37 CAM cell designs in TCAS-I.

with Chen, Li, Taylor, Pedretti

Hardware for Inference & Reasoning

Algorithm–hardware co-design for attention, tree search, and reasoning — IMC-MCTS, Neuro-Symbolic policy hardware, Hamming-binarized attention.

with Chen, Li, HPE Labs (Natarajan, Ignowski)

Selected Publications

[See all publications]

T. Molom-Ochir, B. Morris, M. Horton, et al. "CAMformer: Associative Memory is All You Need." IEEE TCAS-I, 2026.
T. Molom-Ochir, B. Taylor, H. Li, Y. Chen. "Advancements in Content-Addressable Memory (CAM) Circuits: State-of-the-Art, Applications, and Future Directions in the AI Domain." IEEE TCAS-I, 2025.
T. Molom-Ochir, B. Taylor, H. Li, Y. Chen. "MonoSparse-CAM: Efficient Tree Model Processing via Monotonicity and Sparsity in CAMs." ISCAS 2025.
B. Morris, T. Molom-Ochir, et al. "NP-CAM: Efficient and Scalable DNA Classification using NoC-Partitioned CAM Architectures." HPCA 2025.