Call for Papers

Important Dates (AoE)

Paper submissions due: July 11, 2026
Notification to authors: August 5, 2026

Final version due: August 19, 2026 

Overview

Using AI for improving computer systems has seen a significant amount of work both in academia and industry, spanning classic machine learning techniques, Generative AI models, and more recently Agentic AI systems. These approaches offer complementary opportunities: classic ML can learn predictive models and policies from system data, Generative AI can help reason about complex system behavior and assist developers or operators, and agents can observe, plan, and act to optimize systems over time. Together, they create new possibilities for improving scheduling, resource management, debugging, performance tuning, configuration management, cloud operations, databases, networks, storage systems, and other parts of the systems stack.

However, deploying such techniques remains challenging. While many published works in this space focus on solving the underlying learning, reasoning, or planning problems, we observe a gap between these fundamental methods and actually using them to drive real-world impact. This gap is especially important for AI-driven systems that may affect production infrastructure, interact with human operators, rely on imperfect observations, and operate under changing workloads and system conditions. The goal of this workshop is to raise awareness of this gap and to seek understanding of why it occurs. Examples of topics include: feature and prompt stability, reliability, availability, integration into rollout and operations processes, verification, safety guarantees, guardrails for autonomous or semi-autonomous actions, feedback loops introduced by learning or agent behavior, debuggability, observability, and explainability.

We hope to identify these problems and bring together practitioners and academic researchers, both on the production systems and AI sides, to work towards a methodology for capturing these problems in academic research. We believe that starting this conversation between the academic and industrial research communities will facilitate the adoption of AI for Systems research in production systems, and will provide the academic community with access to new research problems that exist in real-world deployments but have received less attention in the academic community.

To this end, we invite three types of papers in the broad area of challenges associated with using machine learning in computer systems: 

• Position papers that explore new challenges and design spaces, 

• Short papers that describe completed or early-stage work, and 

• Abstracts that summarize works published in the past two years. 

We encourage a wide variety of submissions, from success-stories to retrospectives, and are particularly interested in anecdotes and experiences from real-world deployments. We also encourage academic and research submissions that are describing techniques that could be applicable to such deployments, even if they do not describe a real system.

A paper accepted to PACMI would not preclude its future publication at a major conference. Accepted papers will have the option to be included in ACM proceedings. Submissions that are likely to generate vigorous discussion will be favored!

Topics of Interest

Some areas of interest include:

• Research on techniques that expose, address problems with, or are otherwise related to the deployment of ML and data-driven methods in systems, particularly in (but not limited to) the following areas:

  • Feature stability, reliability, or availability

  • ML integration into rollout processes

  • Verification for model outputs 

  • Safety guarantees for ML models and AI agents

  • Control system aspects of ML for systems, such as feedback loops

  • Debuggability and explainability

  • Human-in-the-loop methods and operator-facing AI tools for systems management

• Retrospectives, success stories, or negative results from deployments of ML techniques in computer systems.

• Descriptions of systems where ML was considered but ultimately not adopted.

• Benchmarks that provide industry & academia with the opportunity to capture aspects of deployment of ML for systems.

We define computer systems to broadly include (not limited to): Computer Architecture, Network Systems, Operating Systems, Runtime Systems, Software Systems, Code Modeling, Compilers, Databases, Data Centers, Distributed Systems, Security and Performance Tools. We define ML to broadly include any data-driven methods.

Submission Instructions & Formatting

Papers will be submitted electronically in PDF format via the web submission form (HotCRP).

• Position paper submissions must be no longer than 4 pages, excluding references.

• Short paper submissions must be no longer than 4 pages, excluding references.

• Abstracts of published works must be no longer than 2 pages, excluding references. 

All submissions will be single-blind to allow for going into details without worrying about anonymity. Authors are allowed to post their papers on arXiv or other public forums.

The formatting should follow the same guidelines as SOSP submissions:

Use A4 or US letter paper size, with all text and figures fitting inside a 178 x 229 mm (7 x 9 in) block centered on the page, using two columns separated by 8 mm (0.33 in) of whitespace. Use 10-point font (typeface Times Roman, Linux Libertine, etc.) on 12-point (single-spaced) leading. Graphs and figures should be readable without magnification; they are encouraged to be in color, but should remain readable if printed in grayscale. All pages should be numbered, and references within the paper should be hyperlinked.

Submissions violating these rules may not be considered for publication. We encourage you to upload an early draft of the paper well before the deadline to check if the paper meets the formatting rules.

Most of these rules are automatically applied when using the official SIGPLAN LaTeX or MS Word templates from the ACM.

For Latex, we recommend you use:

\documentclass[sigplan,10pt]{acmart}

\renewcommand\footnotetextcopyrightpermission[1]{}

...

\settopmatter{printfolios=true}

\maketitle

\pagestyle{plain}

...