Use this before writing begins. MLSys exists for work where machine learning and computer
systems constrain each other — the CFP spans systems for ML (training, inference,
serving, compilers, runtimes, specialized hardware, hardware-efficient methods,
benchmarks and tooling) and ML for systems (learned scheduling, LLM-driven hardware
design and system optimization), plus adjacent lanes like federated learning, compound AI
and agent systems, privacy/security for ML, and data preparation (2026 topics list,
verified 2026-07-08). Fit failures here are usually routing failures: strong work aimed
at the wrong reviewer pool.
Ask two questions; MLSys wants both answers to be yes.
request patterns of ML workloads) shape the system design?
ML method or its evaluation?
One yes is a re-routing signal: pure systems novelty exercised on an ML workload belongs
at a general systems venue; pure ML novelty measured for speed belongs at an ML venue.
| Project's center of gravity | Better venue | Why |
|---|---|---|
| New OS/storage/scheduling abstraction, ML is one workload among several | OSDI / SOSP / ATC / EuroSys | Their reviewers reward general-purpose systems depth |
| Network protocol or datacenter fabric for training traffic | NSDI / SIGCOMM | Networking-first reviewer pool |
| Microarchitecture or accelerator design | ASPLOS / ISCA / MICRO | Hardware evaluation norms (simulators, RTL) differ |
| New learning algorithm; efficiency is one experiment | NeurIPS / ICML / ICLR | Algorithmic novelty is the contribution |
| HPC-scale training on supercomputers | SC / HPDC | HPC evaluation culture |
| Mechanism where ML and system constraints genuinely interlock | MLSys | This is the house genre |
| Production ML system's design + deep benchmarks, limited novelty | MLSys industrial track (2026+) | Novelty explicitly not required there |
| Early idea, workshop-scale evidence | An MLSys-adjacent workshop first | Expanded workshop papers can return with chair approval |
because of an ML-aware design decision — not incidentally.
this theorem tight?" or "is this abstraction general?"
the problem (see mlsys-related-work).
access — badge culture is part of this venue's identity.
built [mechanism]; which yields [measured system payoff] at [stated cost]."*
against both literatures.
benchmarks required, novelty not required, company identity may remain visible.
Choosing it for a paper whose real asset is a clever new mechanism undersells the
mechanism; choosing research track for a deployment report gets it rejected for
novelty. Route by the paper's strongest honest asset.
Fill these four lines; blanks are routing information.
ML property exploited: ______________________
System constraint engaged: ______________________
Named mechanism: ______________________
Headline systems metric: ______ on workload ______ vs baseline ______
If "named mechanism" is blank -> industrial track or a workshop.
If "ML property" is blank -> general systems venue.
If "system constraint" is blank -> ML venue.
If the metric line is blank -> not ready for any venue; measure first.
drifted fast (the 2026 list added compound AI/agent systems and LLM-driven hardware
design; assume continued drift).
If the project peaks in March, a systems venue with a nearer deadline may serve the
work better than a seven-month hold.
ML-for-systems is in scope, but the reviewer pool that can judge query optimization
depth sits at database venues. Route to MLSys only if the contribution is the
learning-systems machinery (training-data collection, drift handling, inference
latency budget); route to a DB venue if it is optimizer quality.
paper's spine is the accuracy-preservation insight, ML venues fit; it becomes
MLSys-shaped when the kernel/runtime co-design carries equal weight and the
evaluation leads with latency-quality tradeoffs on real serving stacks — the AWQ
pattern in ../../resources/exemplars/library.md.
only.** Topic fits; evidence does not — simulation-only evaluation of a systems
claim is a recognized weak pattern here. Either add a real-cluster deployment at
modest scale or target a venue where simulation is the norm.
demands measurement depth.
a stale comparison fails the fit test before the novelty question is even reached.
published MLPerf holds benchmark papers to benchmark-committee standards.
[Fit] strong MLSys / MLSys-possible / route elsewhere
[Co-design test] ML->system: yes/no ; system->ML: yes/no
[Track] research / industrial / workshop-first
[Best alternative venue] <venue + reason>
[Contribution compression] observed -> built -> measured -> cost
[Next action] <measurement, mechanism naming, track decision, or reroute>