# Optimizing ML Workload Partitioning between CPUs and CIM Accelerators for Heterogeneous Computing

arXiv: https://arxiv.org/abs/2607.05240 · HTML: https://mlsanity.com/paper/2607.05240/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 40 references that are in the map.
- First posted on arXiv 6 Jul 2026.

## Builds on (most cited first, 24 of 24)

- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning](https://mlsanity.com/paper/2506.09985/) (2025)
- [Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecasting](https://mlsanity.com/paper/2301.00493/) (2023)
- [MultiPath: Multiple Probabilistic Anchor Trajectory Hypotheses for Behavior Prediction](https://mlsanity.com/paper/1910.05449/) (2019)
- [DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning](https://mlsanity.com/paper/2411.04983/) (2024)
- [Motion Transformer with Global Intention Localization and Local Movement Refinement](https://mlsanity.com/paper/2209.13508/) (2022)
- [CoverNet: Multimodal Behavior Prediction using Trajectory Sets](https://mlsanity.com/paper/1911.10298/) (2019)
- [OGBench: Benchmarking Offline Goal-Conditioned RL](https://mlsanity.com/paper/2410.20092/) (2024)
- [MotionDiffuser: Controllable Multi-Agent Motion Prediction using Diffusion](https://mlsanity.com/paper/2306.03083/) (2023)
- [DLow: Diversifying Latent Flows for Diverse Human Motion Prediction](https://mlsanity.com/paper/2003.08386/) (2020)
- [Learning in an Uncertain World: Representing Ambiguity Through Multiple Hypotheses](https://mlsanity.com/paper/1612.00197/) (2016)
- [Overcoming Limitations of Mixture Density Networks: A Sampling and Fitting Framework for Multimodal Future Prediction](https://mlsanity.com/paper/1906.03631/) (2019)
- [LookOut: Diverse Multi-Future Prediction and Planning for Self-Driving](https://mlsanity.com/paper/2101.06547/) (2021)
- [Stochastic Multiple Choice Learning for Training Diverse Deep Ensembles](https://mlsanity.com/paper/1606.07839/) (2016)
- [Forecast-MAE: Self-supervised Pre-training for Motion Forecasting with Masked Autoencoders](https://mlsanity.com/paper/2308.09882/) (2023)
- [ScePT: Scene-consistent, Policy-based Trajectory Predictions for Planning](https://mlsanity.com/paper/2206.13387/) (2022)
- [What Drives Success in Physical Planning with Joint-Embedding Predictive World Models?](https://mlsanity.com/paper/2512.24497/) (2025)
- [Trajectory-wise Multiple Choice Learning for Dynamics Generalization in Reinforcement Learning](https://mlsanity.com/paper/2010.13303/) (2020)
- [Tree-structured Policy Planning with Learned Behavior Models](https://mlsanity.com/paper/2301.11902/) (2023)
- [DISCO Nets: DISsimilarity COefficient Networks](https://mlsanity.com/paper/1606.02556/) (2016)
- [VJEPA: Variational Joint Embedding Predictive Architectures as Probabilistic World Models](https://mlsanity.com/paper/2601.14354/) (2026)
- [Denoising with a Joint-Embedding Predictive Architecture](https://mlsanity.com/paper/2410.03755/) (2024)
- [Continuous Visual Autoregressive Generation via Score Maximization](https://mlsanity.com/paper/2505.07812/) (2025)
- [UWM-JEPA: Predictive World Models That Imagine in Belief Space](https://mlsanity.com/paper/2605.25313/) (2026)

## Abstract (arXiv)

Computing-in-Memory (CIM) accelerators execute Matrix-Vector Multiplications (MVMs) in memory, making them a compelling solution for Machine Learning (ML) workloads. However, existing ML workload partitioning approaches for CIM accelerators do not fully account for Resistive Random Access Memory (RRAM) constraints such as limited memory, high write latency, and limited endurance. They also neglect parallelism, low-level architectural effects, or the Central Processing Unit (CPU) as a complementary compute resource. To address these limitations, we propose an Integer Linear Programming (ILP)-based workload partitioning framework for heterogeneous CPU-CIM systems. It minimizes end-to-end inference latency under RRAM constraints, captures parallelism, and combines empirical profiling with analytical models. Using our framework, heterogeneous CPU-CIM execution achieves speedups of up to 30.9x over CPU-only execution on an edge CPU and 7.3x over a high-performance CPU. A Design Space Exploration (DSE) yields further design insights for future CIM accelerators.