LIMCA: LLM for Automating Analog In-Memory Computing Architecture Design Exploration

2025-03-17

AI Summary

Conventional analog in-memory computing (IMC) architecture design relies heavily on manual effort, suffers from low efficiency, and lacks high-quality, circuit-level netlists. Method: This paper proposes the first large language model (LLM)-driven, fully automated IMC architecture design framework. It operates without human-in-the-loop (NHIL) intervention or domain-expert guidance, integrating zero-shot LLM reasoning, a structured IMC knowledge base, SPICE netlist auto-generation with syntactic and functional verification, and hardware-constrained multi-objective performance modeling. Contribution/Results: The framework enables end-to-end, closed-loop design—from architectural space exploration to verifiable circuit-level outputs. Evaluated on MNIST, synthesized crossbar arrays achieve ≥96% accuracy with ≤3 W power consumption. Design exploration time is significantly reduced, enabling scalable, hardware-aware, customized architecture search.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Natural Language Processing: (Large) Language Models

Application Category

• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Resistive crossbars enabling analog In-Memory Computing (IMC) have emerged as a promising architecture for Deep Neural Network (DNN) acceleration, offering high memory bandwidth and in-situ computation. However, the manual, knowledge-intensive design process and the lack of high-quality circuit netlists have significantly constrained design space exploration and optimization to behavioral system-level tools. In this work, we introduce LIMCA, a novel fine-tune-free Large Language Model (LLM)-driven framework for automating the design and evaluation of IMC crossbar architectures. Unlike traditional approaches, LIMCA employs a No-Human-In-Loop (NHIL) automated pipeline to generate and validate circuit netlists for SPICE simulations, eliminating manual intervention. LIMCA systematically explores the IMC design space by leveraging a structured dataset and LLM-based performance evaluation. Our experimental results on MNIST classification demonstrate that LIMCA successfully generates crossbar designs achieving $geq$96% accuracy while maintaining a power consumption $leq$3W, making this the first work in LLM-assisted IMC design space exploration. Compared to existing frameworks, LIMCA provides an automated, scalable, and hardware-aware solution, reducing design exploration time while ensuring user-constrained performance trade-offs.

Problem

• Automates design of analog In-Memory Computing architectures.
• Generates and validates circuit netlists without manual intervention.
• Explores design space for high accuracy and low power consumption.

Innovation

• LLM-driven framework automates IMC design
• No-Human-In-Loop pipeline for SPICE simulations
• Structured dataset enables systematic design exploration