In-Memory Computing

Follow this topic Ask about this topic

Momentum

2 papers in the last four weeks, against 2 the four weeks before. 0.0% of all new papers.

Jul 13Week of Sep 28

Latest papers 22

All topics

IMC-CLINIC: Coupled Loss-Informed Newton Iterations for Clipping in Analog In-Memory Computing
Sep 28, 2026Yung-Chin Chen, Chia-Yu Chen, Naveen VermaIn-Memory Computing Gradient Clipping
Princeton University, NJ, USA · EnCharge AI, CA, USA
Bio-inspired Learning and Decision-Making with Probabilistic In-Memory Computing Hardware: Part 1
Sep 12, 2026Thomas Dalgaty, Eiji Kawasaki, Miguel de Prado +2Brain-Inspired Synergistic Framework In-Memory Computing
CEA-List, Grenoble, France
RACE-AIMC: Selective Inference for Heterogeneous Analog In-Memory Accelerators at the Edge
Sep 2, 2026Osama Yousuf, Martin Lueker-BodenIn-Memory Computing Neural Network Inference
WD Research San Jose, CA, USA
ASSERT: Adaptive Stochastic Sampling for Robust Diffusion Models on Analog Compute-in-Memory Hardware
Sep 1, 2026Yuannuo Feng, Yizhe Chen, Wenshuai Yao +4Diffusion Models In-Memory Computing
School of Integrated Circuit Science and Engineering, Beihang University, Beijing, China · Zhicun Research Lab, Beijing, China · School of Integrated Circuits, Peking University, Beijing, China +1
NANQ: Noise-Floor-Aware Mixed-Precision Non-Uniform Quantization for Analog Compute-in-Memory
Aug 3, 2026Yizhe Chen, Wenshuai Yao, Saiya Wang +6In-Memory Computing Mixed-Precision Quantization
School of Integrated Circuit Science and Engineering, Beihang University, Beijing, China · School of Integrated Circuits, Peking University, Beijing, China · Department of Electrical and Computer Engineering, The University of Hong Kong, Hong Kong SAR, China
Mitigating the Impact of Retention Loss on Inference Accuracy in 65 nm Single-Poly Floating-Gate Analog In-Memory Computing
Jul 27, 2026Mirko Brazzini, Giulio Filippeschi, Alessandro Catania +2In-Memory Computing Neural Network Inference
Dipartimento di Ingegneria dell’Informazione, University of Pisa, Pisa, Italy · Quantavis SRL, Pisa, Italy
Multi-primitive in-memory computing for Monte Carlo tree search
Jul 24, 2026Tergel Molom-Ochir, Benjamin F. Morris, Yintao He +6In-Memory Computing Monte Carlo Tree Search
Department of Electrical and Computer Engineering, Duke University, Durham, 27708, NC, USA. · Hewlett Packard Labs, Milpitas, 95035, CA, USA.
Leveraging ECRAM for Edge Continual Learning
Jul 22, 2026Nabila Tasnim, Haoran Liu, Qing Cao +1Replay-Based Continual Learning Edge Devices
University of Illinois Urbana-Champaign
NIFA: Nonlinear IMC enhanced FPGA for efficient ML inference
Jul 16, 2026Jiajun Hu, Ruthwik Reddy Sunketa, Lei Zhao +3In-Memory Computing Field-Programmable Gate Arrays
Arizona State University, Tempe, AZ, USA · Hewlett Packard Enterprise Labs, Fort Collins, CO, USA
ExaGEMM: Exploration Framework for CPU-Driven ML Inference via Associative In-Register Computing for Low-Bit GEMM
Jul 16, 2026Hyunwoo Oh, Suyeon Jang, Hanning Chen +3Matrix Multiplication In-Memory Computing
University of California, Irvine
Optimizing ML Workload Partitioning between CPUs and CIM Accelerators for Heterogeneous Computing
Jul 6, 2026Joel Klein, Rebecca Pelke, Roberto Laudani +2Inference Workloads In-Memory Computing
Institute for Communication Technologies and Embedded Systems, RWTH Aachen University, Germany
ReRAM-aware Model Finetuning addressing I-V Non-linearity and Retention Errors
Jun 16, 2026Ching-Yi Lin, Shamik Kundu, Arnab Raha +1In-Memory Computing Neural Network Inference
University of Maryland, College Park, MD, USA · Intel Corporation, 2200 Mission College Blvd, Santa Clara, CA 95054, USA
Neural dynamical systems on ferroelectric compute-in-memory for real-time forecasting
Jun 15, 2026Keshava Katti, Adithya Selvakumar, Pratik Chaudhari +1Neuromorphic Computing Neural Dynamics
Electrical & Systems Engineering, University of Pennsylvania, Philadelphia, PA 19104, USA
Positional Encoding in the Context of Memristor-Based Analog Computation for Automatic Speech Recognition
Jun 11, 2026Benedikt Hilmes, Nick Rossenbach, Ralf SchlüterIn-Memory Computing Reram Crossbar
Machine Learning and Human Language Technology Group, Faculty of Computer Science, RWTH Aachen University, Aachen, Germany
Modern analog computing for solving differential and matrix equations
Jun 11, 2026Zhong Sun, Piergiulio Mannocci, Manuel Le Gallo +1In-Memory Computing Analog
Institute for Artificial Intelligence, School of Integrated Circuits, Peking University, Beijing Advanced Innovation Center for Integrated Circuits, Beijing 100871, China. · 2Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Piazza Leonardo da Vinci 32, Milano, 20133, Italy. · 3IBM Research Europe, Rüschlikon, Switzerland.
CRAM-ER: Error-Resilient Spintronic Computational Random Access Memory for Scalable In-Memory Computation
Jun 1, 2026Sohan Salahuddin Mugdho, Md. Shahedul Hasan, Brahmdutta Dixit +3In-Memory Computing Reram Crossbar
Electrical and Computer Engineering Iowa State University of Science and Technology Ames, IA, USA · Electrical and Computer Engineering University of Minnesota Twin Cities Minneapolis, MN, USA
ROMER: Expert Replacement and Router Calibration for Robust MoE LLMs on Analog Compute-in-Memory Systems
May 12, 2026Wenyong Zhou, Yuannuo Feng, Yizhe Chen +6Mixture-Of-Experts Large Language Models In-Memory Computing
The Department of Electrical and Computer Engineering, The University of Hong Kong, Hong Kong · The School of Integrated Circuit Science and Engineering, Beihang University, Beijing, China
A Controlled Diagnostic Study of Hardware-Induced Distortions in Hardware-Aware Training
May 10, 2026Yunxuan Fang, Xinhe WangHardware-Aware In-Memory Computing
Beihang University Beijing, China
When Stored Evidence Stops Being Usable: Scale-Conditioned Evaluation of Agent Memory
May 8, 2026Jiaqi Shao, Yiyi Lu, Yunzhen Zhang +1Long-Term Agent Memory In-Memory Computing
Hong Kong University of Science and Technology, Hong Kong · Duke Kunshan University, China
Multibit neural inference in a N-ary crossbar architecture
Apr 28, 2026Anatole Moureaux, Anthony Lopes Temporao, Flavio Abreu AraujoIn-Memory Computing Reram Crossbar
Universit´e catholique de Louvain, Institute of Condensed Matter and Nanosciences, Louvain-la-Neuve, 1348, Belgium
Dynamic Symmetric Point Tracking: Tackling Non-ideal Reference in Analog In-memory Training
Feb 24, 2026Quan Xiao, Jindan Li, Zhaoxian Wu +2In-Memory Computing Reram Crossbar
Department of Electrical and Computer Engineering, Cornell University, New York, NY · IBM T. J. Watson Research Center, Yorktown Heights, NY · Rensselaer Polytechnic Institute, Troy, NY
Optimizing Binary and Ternary Neural Network Inference on RRAM Crossbars using CIM-Explorer
May 20, 2025Rebecca Pelke, José Cubero-Cascante, Nils Bosbach +5Reram Crossbar In-Memory Computing
RWTH Aachen University, Germany

Sep 28, 2026, cs.AIJ/K move · Enter open · S save

IMC-CLINIC: Coupled Loss-Informed Newton Iterations for Clipping in Analog In-Memory Computing

Yung-Chin Chen, Chia-Yu Chen, Naveen Verma

Princeton University, NJ, USA · EnCharge AI, CA, USA

In-Memory Computing Gradient Clipping Analog Newton

Open paper AskPDF arXiv

Analog in-memory computing (IMC) offers a promising path toward energy-efficient large language model (LLM) inference by executing matrix multiplications (MatMul) directly within memory arrays in the analog domain. Its efficiency, however, comes with an additional source of error: limited-precision analog-to-digital converters (ADCs) quantize accumulated analog partial sums, introducing output-side error distinct from conventional activation and weight quantization at the MatMul inputs. Clipping can mitigate both operand and ADC quantization errors, but the optimal clipping factors must jointly balance activation rounding and clipping, weight rounding and clipping, and ADC quantization. Existing clipping methods, designed for digital quantization, do not explicitly optimize these coupled sources of IMC error and often rely on costly search-based calibration. We introduce IMC-CLINIC (Coupled Loss-Informed Newton Iterations for Clipping), a clipping calibration framework based on an analytical surrogate for IMC MatMul output error. The surrogate jointly models operand quantization, accumulated clipping-induced bias, and ADC quantization, enabling efficient evaluation of its gradient and approximate curvature from a small calibration set. IMC-CLINIC jointly optimizes activation and weight clipping factors using a safeguarded Newton-type method. Across multiple models and datasets, it improves average zero-shot accuracy by 6.5-11.5 percentage points over the grid search baseline while reducing calibration time by factors of 10.0-12.1. Its analytical surrogate closely tracks empirical IMC output error, and its optimizer is certified within 1% of the global optimum under the loss objective across all projections on two representative models.

Momentum

Related topics

Latest papers 22