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

2026-09-28

AI Summary

This study addresses the accuracy degradation in analog compute-in-memory (CIM) systems caused by the coupling of ADC quantization and operand errors, as well as the inefficiency of conventional clipping calibration. To overcome these limitations, this work proposes IMC-CLINIC, a framework that establishes the first joint analytical surrogate model for multi-source errors. By employing a safeguarded Newton iteration method to replace computationally expensive search procedures, it achieves zero-shot joint optimization of activation and weight clipping factors. This approach effectively decouples and suppresses compounded errors, yielding an average accuracy improvement of 6.5%–11.5% while reducing calibration time by over an order of magnitude. The optimized solutions closely approximate the global optimum, establishing an efficient paradigm for high-accuracy, low-cost CIM deployment.

Abstract

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.

Problem

• Analog In-Memory Computing
• Clipping Calibration
• ADC Quantization Error
• Coupled Error Optimization
• Large Language Model Inference

Innovation

• Analog In-Memory Computing
• Clipping Calibration
• Newton Iterations
• Analytical Surrogate
• ADC Quantization