SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.03644 · NeurIPS · 2024

Unlearnable 3D Point Clouds: Class-wise Transformation Is All You Need

Wei Liu, Hai Jin, Hangtao Zhang, Xianlong Wang, Ziqi Zhou, Minghui Li, Shengshan Hu, Yechao Zhang

arXiv · PDF · Open in the Atlas

Code that ran

17 of 17 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; 2 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 14 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/2410.03644.json · badge: /api/ran/2410.03644.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
CGCL-codes/UnlearnablePC canonical (classifier label)
classifier basis: census
13 ran · 0 no recorded run
cgcl-codes/unlearnablepc canonical (classifier label)
classifier basis: classifier_2025
4 ran · 0 no recorded run
FunctionStatusWhere it lives
build_models Ran
ran; our contract draft was wrong, not the code
cgcl-codes/unlearnablepc/reverse_train.py
code served (permissive licence) · get_code("36efd318d37a6214")
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) and returned MockClassifier. 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
import importlib
import sys
from types import ModuleType

torch.manual_seed(0)

# Create a mock target model module
mock_model = ModuleType("mock_target_model")

class MockClassifier(nn.Module):
    def __init__(self, num_classes, normal_channel=False):
        super().__init__()
        self.num_classes = num_classes
        self.normal_channel = normal_channel
        self.fc = nn.Linear(10, num_classes)
    
    def forward(self, x):
        return self.fc(x)

def get_model(num_classes, normal_channel=False):
    return MockClassifier(num_classes, normal_channel=normal_channel)

mock_model.get_model = get_model

# Register the mock module
sys.modules["mock_target_model"] = mock_model

# Create args object
class Args:
    def __init__(self):
        self.target_model = "mock_target_model"
        self.NUM_CLASSES = 5
        self.device = torch.device("cpu")

ARGS = {
    "args": Args()
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type MockClassifier — a MockClassifier output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a MockClassifier 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
class_wise_transformation Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cgcl-codes/unlearnablepc/utils.py
code served (permissive licence) · get_code("7c11b2cd4e1e7e86")
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) and returned ndarray, ndim 2, dtype family float, shape (4, 3) 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
import torch

torch.manual_seed(0)

# Create a small point cloud: 4 points with 3 coordinates (x, y, z)
point_cloud = torch.randn(4, 3, dtype=torch.float32)

# Define transformation parameters for different modes
# For 'rot' mode: list should have rotation angles [x_deg, y_deg, z_deg] per class
rot_list = {
    0: [10.0, 20.0, 30.0],
    1: [5.0, 15.0, 25.0]
}

# For 'scale' mode: list should have scale factors per class
scale_list = {
    0: 1.5,
    1: 2.0
}

# For 'shear' mode: list should have shear parameters [a, b] per class
shear_list = {
    0: [0.1, 0.2],
    1: [0.15, 0.25]
}

# For 'twist' mode: list should have twist angles per class
twist_list = {
    0: 15.0,
    1: 25.0
}

# For 'taper' mode: list should have taper factors per class
taper_list = {
    0: 5.0,
    1: 10.0
}

# For 'translation' mode: list should have translation vectors [dx, dy, dz] per class
translation_list = {
    0: [0.5, 0.3, 0.2],
    1: [1.0, 0.5, 0.3]
}

# Test with 'rot' mode
ARGS = {
    'point_cloud': point_cloud,
    'mode': 'rot',
    'list': rot_list,
    'label': 0
}
Setup block drafted by
a language model, Claude Haiku 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 ndarray, ndim 2, dtype family float, shape (4, 3) — shape (4, 3) 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
closest_pool Ran
ran; our contract draft was wrong, not the code
CGCL-codes/UnlearnablePC/model_utils/kpconv_util.py
code served (permissive licence) · get_code("728888ce9772c062")
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 Tensor, ndim 2, dtype family float, shape (3, 4) 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 torch

torch.manual_seed(0)

# x: [n1, d] features matrix
# Create a small features matrix with n1=2, d=4
x = torch.randn(2, 4)

# inds: [n2, max_num] indices matrix
# Create indices with n2=3, max_num=2
# Values should be valid indices into x (0 or 1)
# The function only uses inds[:, 0], so we need valid first column indices
inds = torch.tensor([[0, 1], [1, 0], [0, 1]], dtype=torch.long)

ARGS = {
    'x': x,
    'inds': inds
}
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 Tensor, ndim 2, dtype family float, shape (3, 4) — shape (3, 4) 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
data_preprocess Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cgcl-codes/unlearnablepc/UnlearnablePC_generation.py
code served (permissive licence) · get_code("ba9b0ce245485554")
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) and returned tuple. 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 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)

# The function expects data to be a tuple of (points, target)
# points: a tensor that will be indexed and moved to cuda
# target: a tensor where target[:, 0] is extracted, then moved to cuda

# Create small tensors with distinct dimensions
points = torch.randn(2, 4)  # batch_size=2, num_features=4
target = torch.randn(2, 3)  # batch_size=2, num_targets=3

# Construct the data tuple
data = (points, target)

ARGS = {
    "data": data
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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
data_preprocess Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cgcl-codes/unlearnablepc/reverse_train.py
code served (permissive licence) · get_code("b6a64689cf2243bb")
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) and returned tuple. 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 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)

# The function expects data to be a tuple of (points, target)
# points: a tensor that will be moved to cuda
# target: a tensor with shape (batch_size, ...) where target[:, 0] is extracted
#         then moved to cuda and converted to long

# Create small tensors
batch_size = 2
points_dim = 4
target_dim = 3

points = torch.randn(batch_size, points_dim)
target = torch.randint(0, 10, (batch_size, target_dim)).float()

data = (points, target)

ARGS = {
    "data": data
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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
farthest_point_sample Ran
ran; our contract draft was wrong, not the code
CGCL-codes/UnlearnablePC/data_utils/ModelNetDataLoader10.py
pointer only (licence: MIT); the stored copy get_code serves, from Yuzhe-Fu/FlashFPS, is pointer only (licence: NOASSERTION) · get_code("f80066a00e7156a2")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from Yuzhe-Fu/FlashFPS, and GitHub could not classify the licence of that repository (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.

gather Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
CGCL-codes/UnlearnablePC/model_utils/kpconv_util.py
code served (permissive licence) · get_code("06eb9b038422a6ef")
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 Tensor, ndim 4, dtype family float, shape (2, 3, 4, 3) 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 torch

torch.manual_seed(0)

# x: input tensor with shape [N, D_1, ... D_d]
# We'll use a small tensor with N=2 (batch), D_1=4, D_2=3 (feature dimensions)
x = torch.randn(2, 4, 3)

# idx: indexing tensor with shape [n_1, ..., n_m]
# We'll use shape [2, 3] for idx (2D indexing)
idx = torch.tensor([[0, 1, 0], [1, 0, 1]])

# method: choice of implementation (0, 1, or 2)
method = 2

ARGS = {
    "x": x,
    "idx": idx,
    "method": method,
}
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 Tensor, ndim 4, dtype family float, shape (2, 3, 4, 3) — shape (2, 3, 4, 3) 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_graph_feature Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
CGCL-codes/UnlearnablePC/model.py
code served (permissive licence) · get_code("9a8756778dd8234a")
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 Tensor, ndim 4, 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

torch.manual_seed(0)

# Construct a small tensor for x
# Shape: (batch_size, num_dims, num_points)
# batch_size=2, num_dims=4, num_points=3
x = torch.randn(2, 4, 3)

# k parameter for knn
k = 2

# idx parameter (optional, can be None to let function compute it)
idx = None

ARGS = {
    "x": x,
    "k": k,
    "idx": idx
}
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 Tensor, ndim 4, dtype family float — no output shape is stated: observe_signature.py could not observe it (call:AssertionError)
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_nearest_index Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
CGCL-codes/UnlearnablePC/model_utils/gcn3d_util.py
code served (permissive licence) · get_code("6c98740a684d1aad")
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 Tensor, ndim 3, dtype family int, shape (2, 4, 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 torch

torch.manual_seed(0)

# Create small tensors with distinct dimensions for clarity
# target: (bs=2, v1=4, 3)
# source: (bs=2, v2=5, 3)
target = torch.randn(2, 4, 3)
source = torch.randn(2, 5, 3)

ARGS = {
    'target': target,
    'source': source,
}
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 Tensor, ndim 3, dtype family int, shape (2, 4, 1) — shape (2, 4, 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_neighbor_index Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
CGCL-codes/UnlearnablePC/model_utils/gcn3d_util.py
code served (permissive licence) · get_code("e9fc6fa3d035121f")
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 Tensor, ndim 3, dtype family int, shape (2, 4, 2) 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 torch

torch.manual_seed(0)

# Create a small batch of vertices
# Shape: (bs=2, vertice_num=4, 3)
vertices = torch.randn(2, 4, 3)

neighbor_num = 2

ARGS = {
    "vertices": vertices,
    "neighbor_num": neighbor_num,
}
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 Tensor, ndim 3, dtype family int, shape (2, 4, 2) — shape (2, 4, 2) 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
indexing_neighbor Ran
ran; our contract draft was wrong, not the code
CGCL-codes/UnlearnablePC/model_utils/gcn3d_util.py
code served (permissive licence) · get_code("ef9629ff019f3436")
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 Tensor, ndim 4, dtype family float, shape (2, 3, 5, 4) 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 torch

torch.manual_seed(0)

# Create tensor with shape (bs=2, vertice_num=3, dim=4)
tensor = torch.randn(2, 3, 4)

# Create index with shape (bs=2, vertice_num=3, neighbor_num=5)
# Each index value must be valid (0-2) to index into vertice_num dimension
index = torch.randint(0, 3, (2, 3, 5))

ARGS = {
    "tensor": tensor,
    "index": index
}
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 Tensor, ndim 4, dtype family float, shape (2, 3, 5, 4) — shape (2, 3, 5, 4) 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
knn Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
CGCL-codes/UnlearnablePC/model.py
code served (permissive licence) · get_code("cdd0141594039dcb")
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) and returned Tensor, ndim 3, dtype family int, shape (2, 3, 2) 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 torch

torch.manual_seed(0)

# x is used with:
# - transpose(2, 1): expects 3D tensor
# - matmul with transposed version: matrix multiplication
# - sum(x ** 2, dim=1, keepdim=True): sum over dimension 1
# - topk(k=k, dim=-1): get top k indices along last dimension
# 
# Shape should be (batch, features, num_points) based on the transpose operations
# Let's use batch=2, features=4, num_points=3

x = torch.randn(2, 4, 3)

# k is the number of nearest neighbors to find
# Should be less than num_points (3), so k=2
k = 2

ARGS = {
    "x": x,
    "k": k,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family int, shape (2, 3, 2) — shape (2, 3, 2) 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
knn Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
CGCL-codes/UnlearnablePC/model_utils/curvenet_util.py
code served (permissive licence) · get_code("fd43b3106ac73716")
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 Tensor, ndim 3, dtype family int, shape (2, 3, 3) 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 torch

torch.manual_seed(0)

# Create a small tensor for k-nearest neighbors computation
# Shape: (batch_size, num_features, num_points)
# Using distinct dimensions: batch=2, features=4, points=3
x = torch.randn(2, 4, 3)

# k: number of nearest neighbors to return
k = 2

ARGS = {
    "x": x,
    "k": k,
}
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 Tensor, ndim 3, dtype family int, shape (2, 3, 3) — shape (2, 3, 3) 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
normal_knn Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
CGCL-codes/UnlearnablePC/model_utils/curvenet_util.py
code served (permissive licence) · get_code("1979f86a5b2b0446")
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 Tensor, ndim 3, dtype family int, shape (2, 3, 2) 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 torch

torch.manual_seed(0)

# Construct a small tensor x with shape (batch_size, num_features, num_points)
# The function transposes dim 2 and 1, so we need 3 dimensions
# Let's use batch=2, features=4, num_points=3
x = torch.randn(2, 4, 3)

# k is the number of nearest neighbors to find
k = 2

ARGS = {
    "x": x,
    "k": k,
}
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 Tensor, ndim 3, dtype family int, shape (2, 3, 2) — shape (2, 3, 2) 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
pc_normalize Ran
ran, honoured the contract we drafted
CGCL-codes/UnlearnablePC/data_utils/ModelNetDataLoader10.py
pointer only (licence: MIT); the stored copy get_code serves, from ajhamdi/MVTN, is pointer only (licence: NONE) · get_code("4783fbece52f500e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ajhamdi/MVTN, and GitHub detected no licence in that repository when it was checked. 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.

pc_normalize Ran
ran, honoured the contract we drafted
CGCL-codes/UnlearnablePC/model_utils/curvenet_util.py
pointer only (licence: MIT); the stored copy get_code serves, from aboulch/ConvPoint, is pointer only (licence: NOASSERTION) · get_code("ec413739d406e611")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from aboulch/ConvPoint, and GitHub could not classify the licence of that repository (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.

radius_gaussian Ran
ran, violated the contract we drafted
CGCL-codes/UnlearnablePC/model_utils/kpconv_util.py
code served (permissive licence) · get_code("42f2cd40ba483d3b")
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 Tensor, ndim 3, dtype family float, shape (2, 3, 4) 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 torch

torch.manual_seed(0)

# sq_r: input radiuses with shape [dn, ..., d1, d0]
# Creating a tensor with shape [2, 3, 4] to represent multiple radius values
sq_r = torch.tensor([
    [[1.0, 2.0, 3.0, 4.0],
     [0.5, 1.5, 2.5, 3.5],
     [0.1, 0.2, 0.3, 0.4]],
    [[5.0, 6.0, 7.0, 8.0],
     [4.5, 5.5, 6.5, 7.5],
     [0.5, 1.0, 1.5, 2.0]]
])

# sig: extents of gaussians, can be [d1, d0], [d0], or float
# Using shape [3, 4] to match the last two dimensions of sq_r
sig = torch.tensor([[1.0, 2.0, 3.0, 4.0],
                     [0.5, 1.0, 1.5, 2.0],
                     [2.0, 2.5, 3.0, 3.5]])

# eps: small value for numerical stability (optional, has default)
eps = 1e-09

ARGS = {
    'sq_r': sq_r,
    'sig': sig,
    'eps': eps
}
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 Tensor, ndim 3, dtype family float, shape (2, 3, 4) — shape (2, 3, 4) 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

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 12 are also present in other papers whose code we hold (12 with at least one paper dated earlier, 4 the same month, 4 later) and 5 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
knn 52 other papers
48 dated earlier, 2 the same month, 2 later · 12 only as an assembly brick
1801.07829 (2018-01)
Dynamic Graph CNN for Learning on Point Clouds
1905.06292 (2019-05)
3D Point Cloud Generative Adversarial Network Based on Tree Structured Graph Convolutions
2005.06734 (2020-05)
Dense-Resolution Network for Point Cloud Classification and Segmentation
and 49 more not named here
pc_normalize 47 other papers
38 dated earlier, 1 the same month, 7 later, 1 same year, order not determinable
1811.07246 (2018-11)
PointConv: Deep Convolutional Networks on 3D Point Clouds
1912.12098 (2019-12)
Quaternion Equivariant Capsule Networks for 3D Point Clouds
2003.00492 (2020-03)
PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling
and 44 more not named here
farthest_point_sample 43 other papers
35 dated earlier, 1 the same month, 6 later, 1 same year, order not determinable
1811.07246 (2018-11)
PointConv: Deep Convolutional Networks on 3D Point Clouds
2003.00492 (2020-03)
PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling
2010.05272 (2020-10)
IF-Defense: 3D Adversarial Point Cloud Defense via Implicit Function based Restoration
and 40 more not named here
pc_normalize 36 other papers
32 dated earlier, 2 later, 2 same year, order not determinable
1803.05827 (2018-03)
Local Spectral Graph Convolution for Point Set Feature Learning
1904.02375 (2019-04)
ConvPoint: Continuous Convolutions for Point Cloud Processing
1904.10014 (2019-04)
Linked Dynamic Graph CNN: Learning on Point Cloud via Linking Hierarchical Features
and 33 more not named here
get_graph_feature 13 other papers
12 dated earlier, 1 the same month · 2 only as an assembly brick
1801.07829 (2018-01)
Dynamic Graph CNN for Learning on Point Clouds
2109.00179 (2021-09)
Spatio-temporal Self-Supervised Representation Learning for 3D Point Clouds
2111.04426 (2021-11)
3D Siamese Voxel-to-BEV Tracker for Sparse Point Clouds
and 10 more not named here
gather 8 other papers
8 dated earlier
1904.08889 (2019-04)
KPConv: Flexible and Deformable Convolution for Point Clouds
2007.12668 (2020-07)
KPRNet: Improving projection-based LiDAR semantic segmentation
2111.12591 (2021-11)
Lepard: Learning partial point cloud matching in rigid and deformable scenes
and 5 more not named here
radius_gaussian 8 other papers
8 dated earlier
1904.08889 (2019-04)
KPConv: Flexible and Deformable Convolution for Point Clouds
2007.12668 (2020-07)
KPRNet: Improving projection-based LiDAR semantic segmentation
2111.12591 (2021-11)
Lepard: Learning partial point cloud matching in rigid and deformable scenes
and 5 more not named here
closest_pool 6 other papers
6 dated earlier
1904.08889 (2019-04)
KPConv: Flexible and Deformable Convolution for Point Clouds
2007.12668 (2020-07)
KPRNet: Improving projection-based LiDAR semantic segmentation
2111.12591 (2021-11)
Lepard: Learning partial point cloud matching in rigid and deformable scenes
and 3 more not named here
knn 3 other papers
3 dated earlier
2201.12296 (2022-01)
Benchmarking Robustness of 3D Point Cloud Recognition Against Common Corruptions
2203.04041 (2022-03)
Shape-invariant 3D Adversarial Point Clouds
2308.05525 (2023-08)
Robustifying Point Cloud Networks by Refocusing
normal_knn 3 other papers
3 dated earlier
2201.12296 (2022-01)
Benchmarking Robustness of 3D Point Cloud Recognition Against Common Corruptions
2203.04041 (2022-03)
Shape-invariant 3D Adversarial Point Clouds
2308.05525 (2023-08)
Robustifying Point Cloud Networks by Refocusing
get_nearest_index 1 other paper
1 dated earlier
2103.07054 (2021-03)
FS-Net: Fast Shape-based Network for Category-Level 6D Object Pose Estimation with Decoupled Rotation Mechanism
get_neighbor_index 1 other paper
1 dated earlier
2103.07054 (2021-03)
FS-Net: Fast Shape-based Network for Category-Level 6D Object Pose Estimation with Decoupled Rotation Mechanism

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

Traditional unlearnable strategies have been proposed to prevent unauthorized users from training on the 2D image data. With more 3D point cloud data containing sensitivity information, unauthorized usage of this new type data has also become a serious concern. To address this, we propose the first integral unlearnable framework for 3D point clouds including two processes: (i) we propose an unlearnable data protection scheme, involving a class-wise setting established by a categoryadaptive allocation strategy and multi-transformations assigned to samples; (ii) we propose a data restoration scheme that utilizes class-wise inverse matrix transformation, thus enabling authorized-only training for unlearnable data. This restoration process is a practical issue overlooked in most existing unlearnable literature, i.e., even authorized users struggle to gain knowledge from 3D unlearnable data. Both theoretical and empirical results (including 6 datasets, 16 models, and 2 tasks) demonstrate the effectiveness of our proposed unlearnable framework. Our code is available at https://github.com/CGCL-codes/UnlearnablePC * Minghui Li is the corresponding author. 38th Conference on Neural Information Processing Systems (NeurIPS 2024).

For agents

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

get_harvested_code_for_paper("2410.03644")
get_code_for_paper("2410.03644")
have("2410.03644")

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

curl https://syntology.ai/api/ran/2410.03644.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.03644.svg)](https://syntology.ai/paper/2410.03644)

Connect an agent — have() is free.