Call for Papers

What is AlignSys about?

Modern AI systems are increasingly deployed as long-running, interactive systems. In these settings, alignment depends not only on training objectives, but also on system design, feedback mechanisms, monitoring, governance, and human oversight.

AlignSys is a venue for work that studies these questions from both theoretical and practical perspectives.

Tracks

AlignSys supports two submission tracks:

• Theory Track (12 pages): foundations, models, and analysis
• Practice Track (10 pages): systems, deployments, and experience

Page limits apply to the main content only. References and appendices are unlimited. See Tracks for detailed guidance.

Topics of interest

Topics of interest include, but are not limited to:

• Feedback and control in aligned systems
• Post-deployment monitoring and auditing
• Human-in-the-loop and governance-aware design
• Alignment tradeoffs and operational constraints
• Privacy, safety, and observability in AI systems
• Case studies and lessons from real deployments

Submission details

• Submission platform: OpenReview (link to be announced)
• Review model: double-blind
• Authoring tools: LaTeX and Microsoft Word are both welcome
• Submission format: a single anonymized PDF
• Preprints: arXiv and other preprints are allowed

If your submission includes code or supplementary material, please use an anonymous repository during review (e.g., anonymous.4open.science) and include the link in the paper.

Important dates

Questions?

If you are unsure whether your work is a good fit, or have questions about submission, feel free to contact hello@alignsys.org.