SMART-AIMC: Sensitivity-Aware Mixed ADC Resolution with Fine-Grained Tunability for Analog In-Memory Computing
cs.AR, cs.AI
Submitted: 2026-11-01
Updated: 2026-11-01
Code: https://github.com/bmurmann/ADC-survey
License: http://creativecommons.org/licenses/by/4.0/
The gist: Analog In-Memory Computing (IMC) is a promising paradigm for accelerating GEMM-based AI workloads by executing computations directly within the memory array in the analog domain.
Terminology
Abstract
Analog In-Memory Computing (IMC) is a promising paradigm for accelerating GEMM-based AI workloads by executing computations directly within the memory array in the analog domain. While IMC delivers substantial energy and throughput improvements, it can suffer from significant task accuracy degradation due to analog-to-digital converter (ADC) quantization noise, which occurs at the GEMM output and is thus different from weight/activation quantization, which occurs at the GEMM input. Mixed ADC resolution allows IMC modules to choose between high- and low-resolution ADC modes with different energy costs, offering a compelling accuracy-energy trade-off. However, selecting ADC resolutions across IMC units while respecting hardware granularity constraints remains challenging. In this paper, we propose SMART-AIMC, a Sensitivity-aware Mixed ADC Resolution framework with fine-grained Tunability for Analog In-Memory Computing. SMART-AIMC correlates the algorithmic task loss to hardware ADC quantization noise in the form of an ADC Channel Sensitivity (ACS) metric. Using this metric, SMART-AIMC systematically, and in a disciplined manner, allocates ADC resources to critical computations. Furthermore, we introduce a sensitivity-based channel reordering scheme that overcomes the hardware granularity constraints with no runtime overhead and approaches the optimal ADC quantization possible for the model. Evaluated across diverse workloads, SMART-AIMC unlocks a superior accuracy-energy Pareto frontier, providing (1) higher task accuracy at matched energy budgets and (2) fine-grained tuning across efficiency targets, from aggressive energy savings to near-digital inference accuracy. At the latter operating point, SMART-AIMC achieves 6.99 times higher modeled system-level energy efficiency for linear projections than the INT8 digital baseline with negligible accuracy degradation.
Sources
- SliceGPT: Compress Large Language Models by Deleting Rows and Columns
- Analog Foundation Models
- PrefixQuant: Eliminating Outliers by Prefixed Tokens for Large Language Models Quantization
- IMC-CLINIC: Coupled Loss-Informed Newton Iterations for Clipping in Analog In-Memory Computing
- Massive Spikes in LLMs are Bias Vectors: Mechanistic Uncovering and Spike-Free Quantization
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression
- FGMP: Fine-Grained Mixed-Precision Weight and Activation Quantization for Hardware-Accelerated LLM Inference
- SliM-LLM: Salience-Driven Mixed-Precision Quantization for Large Language Models
- SqueezeLLM: Dense-and-Sparse Quantization
- SpinQuant: LLM quantization with learned rotations
- Pointer Sentinel Mixture Models
- Microscaling Data Formats for Deep Learning
- ResQ: Mixed-Precision Quantization of Large Language Models with Low-Rank Residuals
- GLU Variants Improve Transformer
- Llama 2: Open Foundation and Fine-Tuned Chat Models
- Qwen2 Technical Report
- RPTQ: Reorder-based Post-training Quantization for Large Language Models
- MixLLM: LLM Quantization with Global Mixed-precision between Output-features and Highly-efficient System Design
Related papers
- Embedded Arena: Iterative Optimization via Hardware Feedback
- WitCert: Sound Runtime Risk Observability and Gating for KV-Cache Quantization
- SparsePixels: Efficient Convolution for Sparse Data on FPGAs
- Golden Ruler: A Numeric Format Catalog with Bit-Exact Conformance Vectors for FP8, BF16, MXFP4, and Microscaling Formats
- AMMA: A Multi-Chiplet Memory-Centric Architecture for Low-Latency 1M Context Attention Serving
- RFChipAgent: Multi-Agentic AI Flow for Analog/RF Chip Design