SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.20035 · 2024

UltraLight VM-UNet: Parallel Vision Mamba Significantly Reduces Parameters for Skin Lesion Segmentation

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 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; 0 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 4 failures of our instrument; 0 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/2403.20035.json · badge: /api/ran/2403.20035.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
wurenkai/UltraLight-VM-UNet canonical (classifier label)
classifier basis: census
5 ran · 0 no recorded run
FunctionStatusWhere it lives
dataset_normalized Ran
ran, violated the contract we drafted
wurenkai/UltraLight-VM-UNet/loader.py
code served (permissive licence) · get_code("e939c18a5bb9c139")
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 ndarray, ndim 4, dtype family float, shape (2, 4, 3, 1) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 numpy as np

# Create a small test dataset of images
# Shape: (batch_size, height, width, channels) or similar
# The function expects imgs to support:
# - .shape attribute
# - np.std() and np.mean()
# - indexing with [i]
# - np.min() and np.max() on individual images
# - arithmetic operations

batch_size = 2
height = 4
width = 3
channels = 1

# Create a small array of image data with random values
imgs = np.random.rand(batch_size, height, width, channels).astype(np.float32) * 100

ARGS = {
    "imgs": imgs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 4, dtype family float, shape (2, 4, 3, 1) — shape (2, 4, 3, 1) as a later run recorded it
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
get_logger Ran
ran; our contract draft was wrong, not the code
wurenkai/UltraLight-VM-UNet/utils.py
code served (permissive licence) · get_code("23a1f34627822bb3")
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 Logger. 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 os
import tempfile
import logging.handlers

# Create a temporary directory for logs
temp_log_dir = tempfile.mkdtemp()

ARGS = {
    'name': 'test_logger',
    'log_dir': temp_log_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Logger — a Logger output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Logger 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
get_optimizer Ran
ran; our contract draft was wrong, not the code
wurenkai/UltraLight-VM-UNet/utils.py
code served (permissive licence) · get_code("ae1a2dd9d62e7fad")
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 Adam. 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

torch.manual_seed(0)

# Create a minimal model with parameters
class TinyModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

model = TinyModel()

# Create a config class with all required attributes
class Config:
    def __init__(self, opt='Adam'):
        self.opt = opt
        self.lr = 0.001
        self.rho = 0.9
        self.eps = 1e-8
        self.weight_decay = 0.0
        self.lr_decay = 0.0
        self.betas = (0.9, 0.999)
        self.amsgrad = False
        self.lambd = 0.0001
        self.alpha = 0.75
        self.t0 = 1e6
        self.momentum = 0.9
        self.alpha_rmsprop = 0.99
        self.centered = False
        self.etas = (0.5, 1.2)
        self.step_sizes = (1e-6, 50)
        self.dampening = 0.0
        self.nesterov = False

config = Config(opt='Adam')

ARGS = {
    'config': config,
    'model': model,
}
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 Adam — a Adam output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Adam 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
get_scheduler Ran
ran; our contract draft was wrong, not the code
wurenkai/UltraLight-VM-UNet/utils.py
code served (permissive licence) · get_code("dea42821258184c2")
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 CosineAnnealingLR. 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 math

torch.manual_seed(0)

# Create a simple model and optimizer
model = torch.nn.Linear(4, 2)
optimizer = torch.optim.SGD(model.parameters(), lr=0.1)

# Create config class with all necessary attributes
class Config:
    def __init__(self, sch_type):
        self.sch = sch_type
        self.step_size = 10
        self.gamma = 0.1
        self.last_epoch = -1
        self.milestones = [30, 60, 90]
        self.T_max = 100
        self.eta_min = 0.0
        self.mode = 'min'
        self.factor = 0.1
        self.patience = 10
        self.threshold = 1e-4
        self.threshold_mode = 'rel'
        self.cooldown = 0
        self.min_lr = 0.0
        self.eps = 1e-8
        self.T_0 = 10
        self.T_mult = 2
        self.warm_up_epochs = 5
        self.epochs = 100

# Test with one scheduler type
config = Config('CosineAnnealingLR')

ARGS = {
    'config': config,
    'optimizer': optimizer
}
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 CosineAnnealingLR — a CosineAnnealingLR output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a CosineAnnealingLR 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
val_one_epoch Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
wurenkai/UltraLight-VM-UNet/engine.py
code served (permissive licence) · get_code("15d76d94bb1b8c16")
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 float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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 torch.nn as nn
import numpy as np
from unittest.mock import Mock
import logging

torch.manual_seed(0)

# Define a minimal model that satisfies the function's usage
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv = nn.Conv2d(3, 1, kernel_size=1)
    
    def forward(self, x):
        return self.conv(x)

# Create test loader with tiny batches
test_loader = [
    (torch.randn(2, 3, 4, 4), torch.randn(2, 1, 4, 4)),
    (torch.randn(2, 3, 4, 4), torch.randn(2, 1, 4, 4)),
]

# Create model
model = MinimalModel()

# Create criterion
criterion = nn.BCEWithLogitsLoss()

# Create config with required attributes
class Config:
    def __init__(self):
        self.val_interval = 1
        self.threshold = 0.5

config = Config()

# Create logger
logger = logging.getLogger('test_logger')
logger.setLevel(logging.INFO)
handler = logging.StreamHandler()
logger.addHandler(handler)

# Define ARGS
ARGS = {
    'test_loader': test_loader,
    'model': model,
    'criterion': criterion,
    'epoch': 1,
    'logger': logger,
    'config': config,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn, torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 5 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 5 the same month, 3 later) and 0 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
get_logger 6 other papers
3 dated earlier, 1 the same month, 2 later
2307.08473 (2023-07)
EGE-UNet: an Efficient Group Enhanced UNet for skin lesion segmentation
2401.10166 (2024-01)
VMamba: Visual State Space Model
2402.02491 (2024-02)
VM-UNet: Vision Mamba UNet for Medical Image Segmentation
and 3 more not named here
get_optimizer 6 other papers
3 dated earlier, 1 the same month, 2 later
2307.08473 (2023-07)
EGE-UNet: an Efficient Group Enhanced UNet for skin lesion segmentation
2401.10166 (2024-01)
VMamba: Visual State Space Model
2402.02491 (2024-02)
VM-UNet: Vision Mamba UNet for Medical Image Segmentation
and 3 more not named here
get_scheduler 6 other papers
3 dated earlier, 1 the same month, 2 later
2307.08473 (2023-07)
EGE-UNet: an Efficient Group Enhanced UNet for skin lesion segmentation
2401.10166 (2024-01)
VMamba: Visual State Space Model
2402.02491 (2024-02)
VM-UNet: Vision Mamba UNet for Medical Image Segmentation
and 3 more not named here
val_one_epoch 3 other papers
2 dated earlier, 1 the same month
2401.10166 (2024-01)
VMamba: Visual State Space Model
2402.02491 (2024-02)
VM-UNet: Vision Mamba UNet for Medical Image Segmentation
2403.13642 (2024-03)
H-vmunet: High-order Vision Mamba UNet for Medical Image Segmentation
dataset_normalized 1 other paper
1 the same month
2403.13642 (2024-03)
H-vmunet: High-order Vision Mamba UNet for Medical Image Segmentation

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

Traditionally for improving the segmentation performance of models, most approaches prefer to use adding more complex modules. And this is not suitable for the medical field, especially for mobile medical devices, where computationally loaded models are not suitable for real clinical environments due to computational resource constraints. Recently, state-space models (SSMs), represented by Mamba, have become a strong competitor to traditional CNNs and Transformers. In this paper, we deeply explore the key elements of parameter influence in Mamba and propose an UltraLight Vision Mamba UNet (UltraLight VM-UNet) based on this. Specifically, we propose a method for processing features in parallel Vision Mamba, named PVM Layer, which achieves excellent performance with the lowest computational load while keeping the overall number of processing channels constant. We conducted comparisons and ablation experiments with several state-of-the-art lightweight models on three skin lesion public datasets and demonstrated that the UltraLight VM-UNet exhibits the same strong performance competitiveness with parameters of only 0.049M and GFLOPs of 0.060. In addition, this study deeply explores the key elements of parameter influence in Mamba, which will lay a theoretical foundation for Mamba to possibly become a new mainstream module for lightweighting in the future. The code is available from https://github.com/wurenkai/UltraLight-VM-UNet .

For agents

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

get_harvested_code_for_paper("2403.20035")
get_code_for_paper("2403.20035")
have("2403.20035")

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

curl https://syntology.ai/api/ran/2403.20035.json

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

[![Syntology run record](https://syntology.ai/api/ran/2403.20035.svg)](https://syntology.ai/paper/2403.20035)

Connect an agent — have() is free.