LACS 2026

Workshop on Learning-Augmented Compilers & Systems

Held in conjunction with PACT 2026

Monday, October 19, 2026 (Afternoon)

Illini Center, Chicago, Illinois, USA

Overview

Machine learning and AI are increasingly influencing how compilers, runtime systems, and computer architectures are designed, optimized, and used. Learned techniques can replace or complement hand-designed heuristics, guide complex optimization decisions, support adaptive execution, and help researchers and developers explore increasingly large design spaces.

At the same time, AI is changing the workflow through which compiler and systems technologies are developed. Learning-based tools can assist with performance analysis, optimization discovery, autotuning, debugging, and experimental exploration while remaining grounded in compiler analyses, system measurements, and correctness constraints.

The Learning-Augmented Compilers & Systems (LACS) workshop brings together researchers and practitioners working at the intersection of:

LACS provides a forum for sharing ongoing research, early results, practical experiences, and forward-looking ideas about how learning can augment compiler and systems technologies.

Schedule

Time Speakers Affiliation Talk title and abstract
1:30–1:35 PM Opening remarks
1:35–2:05 PM Nikhil Rout,
Blaise Tine
UCLA
TinyVortex: A Learning-Guided Kernel Scheduling Engine for Configurable SIMT GPU Co-Design Space Exploration

Abstract: Tensor compilers typically optimize kernels for fixed hardware targets, overlooking opportunities for joint hardware–software design-space exploration. To this end, we present TinyVortex, an open-source fork of the tinygrad Tensor compiler targeting the configurable RISC-V Vortex GPGPU as the backend. TinyVortex enables end-to-end PyTorch-like Tensor program execution, Vortex C++ kernel generation, and performance evaluation across various GPU microarchitectural configurations across the SIMT hierarchy (threads/warps/issue-width/cores). Building on this infrastructure, we introduce a learning-guided autotuning framework that jointly explores kernel scheduling decisions and hardware parameters, which we currently prototype using cycle-level simulation. Preliminary evaluations demonstrate 1.13x and 2.97x geometric-mean speedups over default kernel rendering schedules for vector addition and SGEMM, respectively, highlighting the potential of configuration-aware compilation for GPU hardware–software co-design.

2:05–2:35 PM Xingjian Ding,
Tejas Ramesh,
Jonathan Ramirez,
Anthony Ramirez,
Henry Valladares,
Shuyin Jiao,
Keren Zhou,
Pengfei Su
UC Merced
George Mason University
UC Merced
UC Merced
UC Merced
NCSU
George Mason University, OpenAI
UC Merced
ProfileWeaver: Cost-Aware Profiler AI Harness for Automated Performance Analysis and Optimization of HPC Applications

Abstract: We study optimizers for constrained layers that consume the gradient at the constrained output and update the underlying parameters directly. For a doubly stochastic routing block parameterized by Sinkhorn iterations, training under such an optimizer leaves the score-to-router backward graph with no registered consumer. The runtime cannot discover this, because which gradients each optimizer consumes is recorded nowhere in the graph. We propose gradient-consumer declarations and dead-Jacobian elimination, which let the runtime skip building backward graphs that nothing will read. On our routing workload, the declaration cuts retained autograd state from 82 tensors to one per router; under torch.compile, the differentiated router remains 2.2× slower than the primal path and still retains its backward state. With the optimizer held identical and parameter updates numerically unchanged, eliminating the dead graph makes a full training step 6.7% faster. The optimizer’s own solves are small and dispatch-bound; batching them across routers and compiling the batched program shifts the lowest-latency execution from CPU to GPU as the number of routers grows, and a two-parameter cost model fitted per execution selects the fastest one at every held-out batch size.

3:00–3:30 PM Coffee break
3:30–4:00 PM Chengyue Wang,
Jason Cong
UCLA
Compiling Layout Transformations for NPUs and FPGAs

Abstract: Efficient execution on NPUs and FPGAs requires reconciling producer and consumer access patterns while sustaining parallelism and avoiding memory-bank conflicts. This talk presents our work on compiling layout transformations, from bit-level stream permutations to closed-form mixed-radix swizzles, with applications to streaming FFTs and NPU data movement.
We then examine the challenges of realizing these transformations through existing HLS tools. We share lessons from using LLM agents to optimize HLS source code and refine generated RTL, highlighting performance limitations and correctness failures. These experiences motivated an MLIR agent dialect for contract-guided RTL refinement that makes compiler-derived goals, permitted transformations, and validation conditions explicit. We discuss how mathematical layout models, deterministic compilation, and LLM agents can work together to produce efficient hardware.

4:00–4:30 PM William Moses UIUC, Google Deepmind
Practical Perspectives on AI-Driven Software Engineering

Abstract: Abstract forthcoming.

4:30–5:00 PM Mary Hall University of Utah
What are the Right Abstractions to Support AI Collaboration on Performance Optimization?

Abstract: Abstract forthcoming.

Venue & Date

Venue: Illini Center (University of Illinois)

Date: Monday, October 19 (afternoon)

Call for Participation

We invite submissions on learning-augmented compilers and systems, including new techniques, empirical studies, work in progress, practical experiences, open challenges, and emerging research directions. We are particularly interested in work that applies machine learning or AI to improve: The themes below illustrate the scope of the workshop rather than define strict boundaries. We welcome interdisciplinary work that combines ideas from compilers, systems, architecture, programming languages, machine learning, and performance engineering. Submissions need not fit neatly into one category. We encourage authors to submit bold, unconventional, and high-risk ideas that could open new directions for learning-augmented compilers and systems.

Topics of Interest

Machine Learning for Compiler Optimization

AI-Augmented Compiler and Systems Workflows

Learning-Augmented Runtime and System Optimization

Learning for ML Systems and Emerging Workloads

Hardware–Software–Learning Co-Design

Data, Infrastructure, Evaluation, and Reproducibility

Emerging Methods and Future Directions

We Particularly Encourage

LACS is intended to promote open technical discussion, including discussion of approaches that did not work as expected. We particularly encourage submissions presenting: A learned method does not need to outperform every existing approach to provide a valuable contribution. Well-designed studies that identify limitations, tradeoffs, failure modes, or conditions under which an approach succeeds or fails are strongly encouraged.

Important Dates

Unless otherwise stated, deadlines are at 11:59 p.m. Anywhere on Earth.

Attend LACS: Presentation Format, Travel Support, and Visas

Presentation Format

LACS strongly encourages in-person presentations to facilitate discussion, networking, and engagement among participants. However, we recognize that travel may not be possible for every presenter.

In exceptional circumstances, we may consider remote presentations via a video-conferencing platform (for example, Zoom or Microsoft Teams). Please contact the workshop organizers as early as possible to discuss this option.

PACT 2026 Travel Support

PACT 2026 is pleased to offer competitive travel support for students and early-career researchers attending the conference and its affiliated workshops, including LACS.

  • Up to USD $800 for participants studying or working in the United States.
  • Up to USD $1,400 for international participants.

Travel support is competitive and subject to available funding. We encourage undergraduate students, graduate students, and early-career researchers—defined as researchers within three years of receiving their Ph.D.—to apply.

Application deadline: September 30, 2026
Apply here: PACT 2026 Travel Support

Visa Information

Attendees who require a visa to travel to the United States should visit the PACT 2026 website for information on requesting a visa-support or invitation letter. Please begin this process well in advance, as visa processing times can vary.

Submission Types

LACS 2026 invites Extended Abstract (up to 4 pages) for presentation at the workshop. Authors are welcome to submit work that has already appeared or is currently under review elsewhere, provided the material can be discussed openly. We equally strongly encourage submissions designed to advocate a particular viewpoint, challenge existing assumptions, or outline a future research agenda.

Submission Guidelines

Additional formatting and presentation instructions will be provided through the submission site.

Submission Link

Submit via EasyChair

Organizers

Program Committee

Previous Workshop Series

LACS continues the legacy of:

MLSH – International Workshop on Machine Learning for Software Hardware Co-Design