C4ML 2026

7th Workshop on Compilers for Machine Learning

📅 Sunday, February 1, 2026 📍 Sydney, Australia (ICC Sydney) 🔗 Co-located with CGO 2026

Scope

Machine learning applications are becoming ubiquitous in large-scale production systems. With that growth and the scaling in data volume and model complexity, the focus on efficiently executing machine learning models has become even greater. The push for increased energy efficiency has led to the emergence of diverse heterogeneous system and accelerator architectures.

In parallel, model complexity and diversity pushed for higher productivity systems, more powerful programming abstractions, type systems, language embeddings, frameworks and libraries. Compilers have historically been the bridge between programmer efficiency and high performance code, allowing the expression of code that remains understandable and productive to port and extend, while producing high-performance code for diverse architectures. As such, compiler techniques have been increasingly incorporated into machine learning frameworks. This goes both ways: given the broadening gap between high-level constructs and hardware accelerators, compilers in machine learning frameworks also emerged as natural clients of machine learning techniques, from domain-specific heuristics to autotuning.

This workshop aims to highlight cutting edge work and research that incorporates compiler techniques and algorithms with optimizing machine learning workloads. The workshop topics span from high-level abstract representations to code generation for accelerators.

Program

The workshop features 10 presentations from leading ML compiler experts from industry and academia.

Opening Remarks

11:00 - 11:10

Session 1: Optimization & Analysis

11:10 - 12:45
Magellan: Autonomous Discovery of Novel Compiler Optimization Heuristics with AlphaEvolve
Hongzheng Chen, Alexander Novikov, Ngân (NV) Vũ, Hanna Alam, Zhiru Zhang, Aiden Grossman, Mircea Trofin, Amir Yazdanbakhsh
Google, Google DeepMind, and Cornell University
Analyzing the complexities of graph compilers for machine learning
Kshitij Jain, Satyam Srivastava
d-Matrix Corporation
XTC: A Research Platform for Optimizing AI Workload Operators
Hugo Pompougnac, Christophe Guillon, Sylvain Noiry, Alban Dutilleul, Guillaume Iooss, Fabrice Rastello
Inria, Univ. Grenoble Alpes, CNRS, Grenoble INP

Lunch

12:45 - 13:45

Session 2: Accelerators & Toolchains

13:45 - 15:30
DeepTools: A Full-Stack Machine Learning Compiler for the IBM Spyre Accelerator
Prasanth Chatarasi, Shubham Jain, Alberto Mannari, Sanchari Sen, Swagath Venkataramani, Viji Srinivasan
IBM Research
From Triton to AMD NPU: Compiler-Driven Kernel Generation with MLIR-AIR
Erwei Wang, Emily Furst, Yiannis Papadopoulos, Aaron Knoll, Mike Chu, Joseph Melber, Stephen Neuendorffer, Samuel Bayliss
AMD and AMD Research
From PyTorch to Calyx: An Open-Source Compiler Toolchain for ML Accelerators
Jiahan Xie, Evan Williams, Adrian Sampson
University of California, Santa Cruz and Cornell University
Defeat the Heap: Zero-Copy Data Movement in AXI4MLIR
Elam Cohavi, Nicolas Bohm Agostini, Jude Haris, Antonino Tumeo, David Kaeli, José Cano
University of Glasgow, Northeastern University, and Pacific Northwest National Laboratory

Break

15:30 - 16:00

Session 3: Code Generation & Kernels

16:00 - 17:20
Library Liberation: Competitive Performance Matmul Through Compiler-composed Nanokernels
Arun Thangamani, Md Asghar Ahmad Shahid, Adam Siemieniuk, Rolf Morel, Renato Golin, Alexander Heinecke
Intel Corporation
Enabling Compiler-Driven Transformation of Attention Variants
Ivan Ho, Kunwar Grover, Tobias Grosser
University of Cambridge and AMD
Systematic Code Generation for ML Computations based on Multi-Dimensional Homomorphisms
Ari Rasch, Richard Schulze
University of Muenster

Closing

17:20

Organizers

Jacques Pienaar
Google
Mehdi Amini
Nvidia
Michael Roberts
Meta
Markus Böck
ETH Zurich

Contact Us

For any questions, please contact us at:
c4ml@googlegroups.com