SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2112.00029 · ICLR · 2022

Pixelated Butterfly: Simple and Efficient Sparse Training for Neural Network Models

Christopher Ré, Zhao Song, Kaizhao Liang, Beidi Chen, Tri Dao, Atri Rudra, Jiaming Yang

arXiv · PDF · Open in the Atlas

Code that ran

4 of 6 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (4 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 0 failures of our instrument; 2 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2112.00029.json · badge: /api/ran/2112.00029.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
huggingface/pytorch_block_sparse infrastructure (classifier label)
classifier basis: classifier_2025
4 ran · 1 no recorded run
HazyResearch/pixelfly marked official in Papers with Code
Papers with Code archive
0 ran · 1 no recorded run
FunctionStatusWhere it lives
BlockSparseLinearFunction Ran
ran on a synthesized input · constructed an object; no result was computed
huggingface/pytorch_block_sparse/pytorch_block_sparse/block_sparse_linear.py
pointer only (licence: NOASSERTION) · get_code("90b5ab602b92ab5f")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, huggingface/pytorch_block_sparse (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

BlockSparseMatrix Ran
ran on a synthesized input · constructed an object; no result was computed
huggingface/pytorch_block_sparse/pytorch_block_sparse/block_sparse_linear.py
pointer only (licence: NOASSERTION) · get_code("c4d9adf80ea69faf")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, huggingface/pytorch_block_sparse (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

BlockSparseMatrixBase Ran
ran on a synthesized input · constructed an object; no result was computed
huggingface/pytorch_block_sparse/pytorch_block_sparse/block_sparse_linear.py
pointer only (licence: NOASSERTION) · get_code("2c4fe31e1554158a")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, huggingface/pytorch_block_sparse (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

BlockSparseMatrixEmulator Ran
ran on a synthesized input · constructed an object; no result was computed
huggingface/pytorch_block_sparse/pytorch_block_sparse/block_sparse_linear.py
pointer only (licence: NOASSERTION) · get_code("ed2fbbe21934c400")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, huggingface/pytorch_block_sparse (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

BlockSparseLinear Not yet run
unverified (harvested, no recorded run)
huggingface/pytorch_block_sparse/pytorch_block_sparse/block_sparse_linear.py
pointer only (licence: NOASSERTION) · get_code("bcb76e709020c908")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, huggingface/pytorch_block_sparse (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

BlockdiagButterflyMultiply Not yet run
unverified (harvested, no recorded run)
HazyResearch/pixelfly/src/models/layers/blockdiag_butterfly_multiply.py
code served (permissive licence) · get_code("1b4982391e4c1bf4")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 6 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Overparameterized neural networks generalize well but are expensive to train. Ideally, one would like to reduce their computational cost while retaining their generalization benefits. Sparse model training is a simple and promising approach to achieve this, but there remain challenges as existing methods struggle with accuracy loss, slow training runtime, or difficulty in sparsifying all model components. The core problem is that searching for a sparsity mask over a discrete set of sparse matrices is difficult and expensive. To address this, our main insight is to optimize over a continuous superset of sparse matrices with a fixed structure known as products of butterfly matrices. As butterfly matrices are not hardware efficient, we propose simple variants of butterfly (block and flat) to take advantage of modern hardware. Our method (Pixelated Butterfly) uses a simple fixed sparsity pattern based on flat block butterfly and low-rank matrices to sparsify most network layers (e.g., attention, MLP). We empirically validate that Pixelated Butterfly is 3× faster than butterfly and speeds up training to achieve favorable accuracy-efficiency tradeoffs. On the ImageNet classification and WikiText-103 language modeling tasks, our sparse models train up to 2.5× faster than the dense MLP-Mixer, Vision Transformer, and GPT-2 medium with no drop in accuracy. * Equal contribution. Order determined by coin flip.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2112.00029")
get_code_for_paper("2112.00029")
have("2112.00029")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2112.00029.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2112.00029.svg)](https://syntology.ai/paper/2112.00029)

Connect an agent — have() is free.