SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2409.03460 · 2024

LowFormer: Hardware Efficient Design for Convolutional Transformer Backbones

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 (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 3 failures of our instrument; 2 no recorded run), measured 2026-11-21; 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/2409.03460.json · badge: /api/ran/2409.03460.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
altair199797/lowformer canonical (classifier label)
classifier basis: census
4 ran · 2 no recorded run
FunctionStatusWhere it lives
accuracy Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
altair199797/lowformer/eval_cls_model.py
code served (permissive licence) · get_code("3f7b732a77ba46cf")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create tiny tensors for testing
batch_size = 2
num_classes = 4

# output: logits/scores from model, shape (batch_size, num_classes)
output = torch.randn(batch_size, num_classes)

# target: ground truth class indices, shape (batch_size,)
target = torch.tensor([1, 3])

# topk: tuple of k values to compute accuracy for
topk = (1, 2)

ARGS = {
    "output": output,
    "target": target,
    "topk": topk,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
build_kwargs_from_config Ran
ran; our contract draft was wrong, not the code
altair199797/lowformer/lowformer_model.py
code served (permissive licence) · get_code("f7db1a3ffb0d1114")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from inspect import signature

torch.manual_seed(0)

# Define a simple target function for testing
def sample_model_forward(x, y=None, learning_rate=0.01, verbose=False):
    """A sample function that will be used as target_func"""
    return x + y if y is not None else x

# Create config dict with some parameters that match and some that don't
config = {
    "x": torch.randn(2, 4),
    "y": torch.randn(2, 4),
    "learning_rate": 0.001,
    "verbose": True,
    "nonexistent_param": "should_be_ignored",
    "another_invalid": 42
}

# The target_func parameter
target_func = sample_model_forward

ARGS = {
    "config": config,
    "target_func": target_func
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
val2list Ran
ran, honoured the contract we drafted
altair199797/lowformer/lowformer_model.py
code served (permissive licence) · get_code("f1073b4e93f4484c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned list, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Test case 1: x is a list
x_list = [1, 2, 3]
repeat_time_1 = 1

# Test case 2: x is a tuple
x_tuple = (4, 5, 6)
repeat_time_2 = 1

# Test case 3: x is a single value (not list or tuple) with repeat_time > 1
x_single = 42
repeat_time_3 = 5

# Test case 4: x is a tensor (single value case)
x_tensor = torch.tensor([1.0, 2.0])
repeat_time_4 = 3

# We'll use test case 3 as the main example since it exercises the repeat logic
ARGS = {
    'x': x_single,
    'repeat_time': repeat_time_3
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family int — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
val2tuple Ran
ran; our contract draft was wrong, not the code
altair199797/lowformer/lowformer_model.py
code served (permissive licence) · get_code("7a8e24c5db77ad97")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned tuple, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Test case 1: x is a single value (not list or tuple)
x_single = 5
min_len_single = 3
idx_repeat_single = -1

# Test case 2: x is a list
x_list = [1, 2]
min_len_list = 5
idx_repeat_list = -1

# Test case 3: x is a tuple
x_tuple = (10, 20, 30)
min_len_tuple = 4
idx_repeat_tuple = 1

# Test case 4: x is empty list
x_empty = []
min_len_empty = 2
idx_repeat_empty = -1

# Using test case 2 for the ARGS dict
ARGS = {
    'x': x_list,
    'min_len': min_len_list,
    'idx_repeat': idx_repeat_list
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family int — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
reset_bn Not yet run
unverified (harvested, no recorded run)
altair199797/lowformer/lowformer/models/nn/norm.py
code served (permissive licence) · get_code("7cb8833e8355fe85")
How it ran

no invocation recorded.

testrun_it_old Not yet run
unverified (harvested, no recorded run)
altair199797/lowformer/eval_cls_model.py
code served (permissive licence) · get_code("4d5c65845e67bb73")
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, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 2 later) and 3 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.

FunctionAlso present inEarliest-datedOthers named
val2list 2 other papers
1 dated earlier, 1 later
2304.02643 (2023-04)
Segment Anything
2506.00045 (2025-06)
ACE-Step: A Step Towards Music Generation Foundation Model
val2tuple 2 other papers
1 dated earlier, 1 later
2304.02643 (2023-04)
Segment Anything
2506.00045 (2025-06)
ACE-Step: A Step Towards Music Generation Foundation Model
accuracy 1 other paper
1 dated earlier
2205.14756 (2022-05)
EfficientViT: Multi-Scale Linear Attention for High-Resolution Dense Prediction

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 2026-11-21.

Repositories linked to this paper

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

Abstract

Research in efficient vision backbones is evolving into models that are a mixture of convolutions and transformer blocks. A smart combination of both, architecture-wise and component-wise is mandatory to excel in the speedaccuracy trade-off. Most publications focus on maximizing accuracy and utilize MACs (multiply accumulate operations) as an efficiency metric. The latter however often do not measure accurately how fast a model actually is due to factors like memory access cost and degree of parallelism. We analyzed common modules and architectural design choices for backbones not in terms of MACs, but rather in actual throughput and latency, as the combination of the latter two is a better representation of the efficiency of models in real applications. We applied the conclusions taken from that analysis to create a recipe for increasing hardware-efficiency in macro design. Additionally we introduce a simple slimmed-down version of MultiHead Self-Attention, that aligns with our analysis. We combine both macro and micro design to create a new family of hardware-efficient backbone networks called LowFormer. LowFormer achieves a remarkable speedup in terms of throughput and latency, while achieving similar or better accuracy than current state-of-the-art efficient backbones. In order to prove the generalizability of our hardware-efficient design, we evaluate our method on GPU, mobile GPU and ARM CPU. We further show that the downstream tasks object detection and semantic segmentation profit from our hardware-efficient architecture. Code and models are available at https://github.com/ altair199797/LowFormer.

For agents

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

get_harvested_code_for_paper("2409.03460")
get_code_for_paper("2409.03460")
have("2409.03460")

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

curl https://syntology.ai/api/ran/2409.03460.json

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

[![Syntology run record](https://syntology.ai/api/ran/2409.03460.svg)](https://syntology.ai/paper/2409.03460)

Connect an agent — have() is free.