..  Licensed to the Apache Software Foundation (ASF) under one
    or more contributor license agreements.  See the NOTICE file
    distributed with this work for additional information
    regarding copyright ownership.  The ASF licenses this file
    to you under the Apache License, Version 2.0 (the
    "License"); you may not use this file except in compliance
    with the License.  You may obtain a copy of the License at

..    http://www.apache.org/licenses/LICENSE-2.0

..  Unless required by applicable law or agreed to in writing,
    software distributed under the License is distributed on an
    "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
    KIND, either express or implied.  See the License for the
    specific language governing permissions and limitations
    under the License.

Publications
============

TVM is developed as part of peer-reviewed research in machine learning compiler
framework for CPUs, GPUs, and machine learning accelerators.

This document includes references to publications describing the research,
results, and design that use or built on top of TVM.

2018

* `TVM: An Automated End-to-End Optimizing Compiler for Deep Learning`__, [Slides_]

.. __: https://arxiv.org/abs/1802.04799
.. _Slides: https://www.usenix.org/system/files/osdi18-chen.pdf

* `Learning to Optimize Tensor Programs`__, [Slides]

.. __: https://arxiv.org/pdf/1805.08166.pdf

2020

* `Ansor: Generating High-Performance Tensor Programs for Deep Learning`__, [Slides__] [Tutorial__]

.. __: https://arxiv.org/abs/2006.06762
.. __: https://www.usenix.org/sites/default/files/conference/protected-files/osdi20_slides_zheng.pdf
.. __: https://tvm.apache.org/2021/03/03/intro-auto-scheduler

2021

* `Nimble: Efficiently Compiling Dynamic Neural Networks for Model Inference`__, [Slides__]

.. __: https://arxiv.org/abs/2006.03031
.. __: https://shenhaichen.com/slides/nimble_mlsys.pdf

* `Cortex: A Compiler for Recursive Deep Learning Models`__, [Slides__]

.. __: https://arxiv.org/pdf/2011.01383.pdf
.. __: https://mlsys.org/media/mlsys-2021/Slides/1507.pdf

* `UNIT: Unifying Tensorized Instruction Compilation`__, [Slides]

.. __: https://arxiv.org/abs/2101.08458

* `Lorien: Efficient Deep Learning Workloads Delivery`__, [Slides]

.. __: https://assets.amazon.science/c2/46/2481c9064a8bbaebcf389dd5ad75/lorien-efficient-deep-learning-workloads-delivery.pdf


* `Bring Your Own Codegen to Deep Learning Compiler`__, [Slides] [Tutorial__]

.. __: https://arxiv.org/abs/2105.03215
.. __: https://tvm.apache.org/2020/07/15/how-to-bring-your-own-codegen-to-tvm

2022

* `DietCode: Automatic optimization for dynamic tensor program`__, [Slides]

.. __: https://proceedings.mlsys.org/paper/2022/file/fa7cdfad1a5aaf8370ebeda47a1ff1c3-Paper.pdf

* `Bolt: Bridging the Gap between Auto-tuners and Hardware-native Performance`__, [Slides]

.. __: https://proceedings.mlsys.org/paper/2022/file/38b3eff8baf56627478ec76a704e9b52-Paper.pdf

* `The CoRa Tensor Compiler: Compilation for Ragged Tensors with Minimal Padding`__, [Slides]

.. __: https://arxiv.org/abs/2110.10221