Skip to yearly menu bar Skip to main content


MLSys 2025 Call For Papers

Authors are encouraged to submit previously unpublished research at the intersection of machine learning and computer systems. The MLSys Program Committee will select papers based on a combination of novelty, quality, interest, and impact.

Topics of interest cover both systems for machine learning and machine learning for systems, and include, but are not limited to:

  • Efficient model training, inference, and serving
  • Large language model (LLM) training, fine-tuning, and inference 
  • Compound AI systems and AI agent systems
  • Distributed and federated learning algorithms
  • Privacy and security for ML applications
  • ML methods for job scheduling in computing systems
  • Testing, debugging, and monitoring of ML applications
  • Fairness, interpretability, and explainability for ML applications
  • Data preparation and data cleaning
  • ML programming models and abstractions
  • Programming languages for machine learning
  • ML compilers and runtimes
  • Visualization of data, models, and predictions
  • Specialized hardware for machine learning
  • LLM-based hardware design or system optimization techniques 
  • Hardware-efficient ML methods
  • Machine learning benchmarks, datasets, and tooling

Reviewing process: All submissions will be double-blind, though authors are allowed to post their papers on arXiv or other public forums. Key dates related to the reviewing process are given below:

Dual submission policy: We will not accept any paper which, at the time of submission, is under review for another conference or has already been published. This policy also applies to papers that overlap substantially in technical content with conference papers under review or previously published. After submission and during the review period, MLSys submissions must not be submitted to other conferences. This policy has the following exceptions.

  • Authors may submit to MLSys substantially different versions of journal papers that are currently under review by a journal, but not yet accepted at the time of submission.
  • Authors may submit in parallel to MLSys and to non-archival venues, such as conferences or workshops without proceedings.
  • Authors may post technical reports (e.g., arXiv) of papers submitted to MLSys.
  • Subject to approval by the program chairs, authors may submit expanded versions of papers published at workshops with proceedings. The authors should explain the differences between the workshop version and the full paper and email the workshop paper to the program chairs. These latter submissions will be evaluated on the added novelty compared to their workshop version.

Submission format: To prepare your submission to MLSys 2025, please use the LaTeX style files provided at: mlsys2025style.zip (we are using the same submission format as 2024). Submitted papers will be in a 2-column format and can be up to 10 pages long, not including references. Each reference must explicitly list all authors of the paper. Authors may use as many pages of appendices as they wish, but reviewers are not required to read these. If you wish to include an appendix, please upload this separately from your main paper. You will be able to upload the appendix after submitting your main paper. The appendix is due at the same time as the main paper.

Artifacts: Our conference promotes reproducibility of experimental results and encourages code and data sharing to help the community quickly validate and compare alternative approaches. We invite authors of accepted papers MLSys 2025 to submit their supporting materials (code, data, models, experimental workflows, results) to the Artifact Evaluation process based on the ACM Artifact Review and Badging policy, a standard for systems conferences including ASPLOS, CGO, PLDI, PPoPP and Super Computing. This submission is voluntary and will not influence the final decision regarding the papers.

Conflicts of Interest: Authors must register all their conflicts on the paper submission site. Conflicts are needed to ensure appropriate assignment of reviewers. If a paper is found to have an undeclared conflict that causes a problem OR if a paper is found to declare false conflicts in order to abuse or “game” the review system, the paper may be rejected. Please review the definitions of Conflict of Interest.

Co-author registration: Anyone who plans to submit a paper as an author or a co-author will need to create (or update) their OpenReview profile and be added to the author list of the submission by the full paper submission deadline.

Plagiarism: All forms of plagiarism are disallowed at MLSys, as is any unethical attempt to subvert or corrupt the peer review process. While authors are free to consult a large-scale language model (LLM) such as ChatGPT while writing the paper, papers that include content generated by LLMs or other generative models are prohibited unless this content is illustrative or is presented as a part of the paper’s experimental analysis.

Submissions that violate these instructions may be returned without review, at the discretion of the Program Chairs, to ensure a review process that is fair to all potential authors.