LACS 2026

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:

• Machine learning and artificial intelligence
• Compilers and programming systems
• Runtime and operating systems
• Computer architecture and hardware
• Performance engineering and optimization

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.

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:

• Compiler and system decisions
• Optimization and development workflows
• Runtime adaptation
• Performance analysis
• Hardware–software co-design
• The infrastructure and methodology needed to evaluate learning-augmented systems

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

• Learned cost models and optimization heuristics
• Optimization phase ordering and pass selection
• Register allocation and instruction scheduling
• Search-space exploration and autotuning
• ML-guided optimization in LLVM, MLIR, GCC, TVM & related

AI-Augmented Compiler and Systems Workflows

• AI-assisted optimization discovery & design-space exploration
• Automated kernel generation, optimization, and tuning
• Agent-based compiler and systems optimization
• Compiler debugging, testing, and validation
• Tools for compiler and systems developer productivity

Learning-Augmented Runtime and System Optimization

• Runtime optimization and adaptive execution
• Task scheduling and resource allocation
• Memory management and hierarchy optimization
• Dynamic compilation and runtime specialization
• Energy-, thermal-, and power-aware optimization

Learning for ML Systems and Emerging Workloads

• Compilation and optimization of ML workloads
• Kernel selection, fusion, and scheduling
• Quantization, sparsity, and mixed-precision execution
• Optimization for large models and generative-AI
• Performance portability across heterogeneous architectures

Hardware–Software–Learning Co-Design

• ML-guided computer architecture design
• Compiler–architecture co-optimization
• Design-space exploration for accelerators
• Mapping and scheduling for heterogeneous hardware
• Learning-assisted FPGA, ASIC, and accelerator design

Data, Infrastructure, Evaluation, and Reproducibility

• Training-data generation and curation
• Benchmark suites and shared datasets
• Evaluation methodologies for learned optimizations
• Reproducibility and replication studies
• Generalization across workloads and architectures

Emerging Methods and Future Directions

• Reinforcement learning and sequential decision-making
• Foundation models for compiler and systems optimization
• Explainable and interpretable learned decisions
• Few-shot, transfer, continual, and online learning
• Self-improving compilers and adaptive systems

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:

• Negative results and carefully analyzed failed approaches
• Lessons learned from research or deployment
• Unexpected empirical observations
• Reproducibility and replication studies
• Comparisons showing when traditional heuristics remain preferable
• Industrial experiences and deployment challenges
• Preliminary results and work in progress
• New datasets, benchmarks, tools, and evaluation methodologies
• Open research questions and community challenges
• Bold or unconventional ideas that challenge existing assumptions

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

• Submission Deadline: Friday, August 14, 2026
• Author Notification: Friday, August 28, 2026
• Workshop: Monday, October 19, 2026

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

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.

Organizers

• Eun Jung (EJ) Park, Qualcomm Inc.
• Riyadh Baghdadi, NYU Abu Dhabi & MIT
• Joseph Manzano, Pacific Northwest National Laboratory
• Keren Zhou, George Mason University & OpenAI

Program Committee

TBD