SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2409.19472 · NeurIPS · 2024

Towards Croppable Implicit Neural Representations

Eran Treister, Maor Ashkenazi

arXiv · PDF · Open in the Atlas

Code that ran

13 of 22 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 5 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 6 failures of our instrument; 9 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/2409.19472.json · badge: /api/ran/2409.19472.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
maorash/Local-Global-INRs canonical (classifier label)
classifier basis: census
11 ran · 1 no recorded run
maorash/local-global-inrs canonical (classifier label)
classifier basis: classifier_2025
2 ran · 8 no recorded run
FunctionStatusWhere it lives
BatchLinear Ran
ran on a synthesized input
maorash/local-global-inrs/modules.py
code served (permissive licence) · get_code("61a0eddac98610bc")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 3, 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
from torch import nn

class BatchLinear(nn.Linear):
    __doc__ = nn.Linear.__doc__

    def forward(self, input):
        output = input.matmul(self.weight.permute(*[i for i in range(len(self.weight.shape) - 2)], -1, -2))
        output += self.bias.unsqueeze(-2)
        return output


INIT_ARGS = {
    'in_features': 4,
    'out_features': 3,
    'bias': True
}

ARGS = {
    'input': torch.randn(2, 5, 4)
}

if __name__ == '__main__':
    model = BatchLinear(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Input shape: {ARGS['input'].shape}")
    print(f"Output shape: {output.shape}")
    print(f"Weight shape: {model.weight.shape}")
    print(f"Bias shape: {model.bias.shape}")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
LocallyConnectedLayer Ran
ran on a synthesized input · constructed an object; no result was computed
maorash/local-global-inrs/modules.py
code served (permissive licence) · get_code("ddcaba8b880b8459")
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 an instance of its own class, LocallyConnectedLayer: the run constructed the object and did not call it, so no result was computed. 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
from torch import nn
import math

torch.manual_seed(0)

# The LocallyConnectedLayer expects:
# - in_features: int, number of input features per group
# - out_features: int, number of output features per group
# - groups: int, number of groups
# - bias: bool (optional, defaults to True)

# Looking at the forward method:
# - x is passed to torch.bmm(x, self.weight)
# - self.weight has shape (groups, in_features // groups, out_features // groups)
# - torch.bmm expects (batch, n, m) @ (batch, m, p) -> (batch, n, p)
# - So x should have shape (groups, in_features // groups, something)

# Let's create valid arguments:
in_features = 8
out_features = 6
groups = 2
bias = True

ARGS = {
    'in_features': in_features,
    'out_features': out_features,
    'groups': groups,
    'bias': bias
}
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 LocallyConnectedLayer — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a LocallyConnectedLayer 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
compute_global_hidden_dim Ran
ran, honoured the contract we drafted
maorash/Local-Global-INRs/compute_partitions.py
code served (permissive licence) · get_code("fc4bfe37fd4bad7c")
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 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)

# Analyze the function compute_global_hidden_dim:
# - total_param_number: int, used in multiplication
# - global_weights_factor: float/int, used in multiplication
# - in_dim: int, passed to binary_search and compute_global_weights
# - hidden_layers: int, passed to binary_search and compute_global_weights

ARGS = {
    "total_param_number": 10000,
    "global_weights_factor": 0.3,
    "in_dim": 64,
    "hidden_layers": 3,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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
compute_hidden_dims Ran
ran, honoured the contract we drafted
maorash/Local-Global-INRs/compute_partitions.py
code served (permissive licence) · get_code("be4643ad127966f5")
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 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 numpy as np

ARGS = {
    "total_param_number": 10000,
    "global_weights_factor": 0.3,
    "in_dim": 64,
    "output_dim": 32,
    "hidden_layers": 2,
    "groups": 4,
}
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 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
compute_local_hidden_dim Ran
ran, honoured the contract we drafted
maorash/Local-Global-INRs/compute_partitions.py
code served (permissive licence) · get_code("efa44d56279c6bdf")
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 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)

ARGS = {
    "in_dim": 4,
    "output_dim": 3,
    "hidden_layers": 2,
    "groups": 2,
    "target_param_number": 500
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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
divergence Ran
ran; our contract draft was wrong, not the code
maorash/Local-Global-INRs/utils.py
code served (permissive licence) · get_code("87e822f4bcfbdac7")
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 (2, 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)

# The divergence function computes the divergence of a vector field y with respect to x
# It requires:
# - y: a tensor with shape [..., n] where n is the output dimension
# - x: a tensor that y depends on, requires grad

# Create a small input tensor x with requires_grad=True
x = torch.randn(2, 4, requires_grad=True)

# Create y as a function of x
# y should have shape [..., output_dim]
# We'll create y with output dimension 3 (distinct from input dim 4)
# y = some function of x, e.g., a simple linear-like transformation
y = torch.sum(x, dim=1, keepdim=True) * torch.ones(2, 3)
y.requires_grad_(True)

# Alternatively, if y needs to be explicitly computed from x:
# Let's make y depend on x through autograd graph
y = x[:, :3] * 2.0 + x[:, 1:4].sum(dim=1, keepdim=True)

ARGS = {
    "y": y,
    "x": x,
}
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 (2, 1) — shape (2, 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
function_mse Ran
ran; our contract draft was wrong, not the code
maorash/Local-Global-INRs/loss_functions.py
code served (permissive licence) · get_code("08039d2c26fb8e0f")
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
torch.manual_seed(0)

# Analyze the function:
# - model_output: dict with key 'model_out' containing a tensor
# - gt: dict with key 'func' containing a tensor
# - These tensors are subtracted and squared, then mean is taken
# - Both should have compatible shapes for broadcasting

# Create small tensors (batch=2, dim=4)
model_output_tensor = torch.randn(2, 4)
gt_tensor = torch.randn(2, 4)

model_output = {
    'model_out': model_output_tensor
}

gt = {
    'func': gt_tensor
}

ARGS = {
    'model_output': model_output,
    'gt': gt
}
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
get_mgrid Ran
ran, honoured the contract we drafted
maorash/Local-Global-INRs/dataio.py
code served (permissive licence) · get_code("c3c376e19f06a842")
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 (256, 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
import numpy as np

torch.manual_seed(0)

ARGS = {
    "sidelen": 16,
    "dim": 2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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 (256, 2) — shape (256, 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
get_mgrid Ran
ran, honoured the contract we drafted
maorash/Local-Global-INRs/incode_experiments/dataio.py
code served (permissive licence) · get_code("386eb2e1d4237e59")
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 (256, 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 numpy as np
import torch

torch.manual_seed(0)

ARGS = {
    'sidelen': 16,
    'dim': 2,
    'zero_to_one': False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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 (256, 2) — shape (256, 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
gradient Ran
ran; our contract draft was wrong, not the code
maorash/Local-Global-INRs/utils.py
code served (permissive licence) · get_code("d5a66d1f869bb91d")
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 (2, 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
import torch.nn as nn

torch.manual_seed(0)

# Create a simple computational graph
# We need y to depend on x so that gradients can be computed

# Create input tensor x that requires gradients
x = torch.randn(2, 4, requires_grad=True)

# Create a simple function: y = sum(x^2)
# This creates a computational graph where y depends on x
y = (x ** 2).sum()

# grad_outputs is optional, defaults to ones_like(y)
grad_outputs = torch.ones_like(y)

ARGS = {
    "y": y,
    "x": x,
    "grad_outputs": grad_outputs,
}
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 (2, 4) — shape (2, 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
image_mse Ran
ran; our contract draft was wrong, not the code
maorash/Local-Global-INRs/loss_functions.py
code served (permissive licence) · get_code("4c1c1fa3b72b2ac7")
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

torch.manual_seed(0)

# Create minimal model_output dict
# The function accesses model_output['model_out']
model_output = {
    'model_out': torch.randn(2, 4)
}

# Create minimal gt dict
# The function accesses gt['img'] and optionally gt['mask']
gt = {
    'img': torch.randn(2, 4),
    'mask': torch.ones(2, 4)
}

# mask can be None or a tensor
# When provided as parameter, it overrides gt['mask']
mask = None

ARGS = {
    'mask': mask,
    'model_output': model_output,
    'gt': gt
}
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
laplace Ran
ran; our contract draft was wrong, not the code
maorash/Local-Global-INRs/utils.py
code served (permissive licence) · get_code("a2be4be6558869af")
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 (2, 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 a small tensor y that depends on x via autograd graph
# y needs to have a last dimension so we can iterate over it in divergence()
# Shape: (batch=2, feature_dim=4)
x = torch.randn(2, 4, dtype=torch.float32, requires_grad=True)

# Create y as a function of x so the autograd graph is built
# y should also have shape compatible with the divergence function
# We'll make y = x^2 to create a simple dependency
y = (x ** 2).sum(dim=1, keepdim=True)  # Shape: (2, 1)

ARGS = {
    "y": y,
    "x": x,
}
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 (2, 1) — shape (2, 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
lin2img Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
maorash/Local-Global-INRs/dataio.py
code served (permissive licence) · get_code("3e90381aa457b40b")
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, 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 numpy as np
import torch

torch.manual_seed(0)

# Case 1: Without slices - simple linear to image reshape
tensor_no_slices = torch.randn(2, 16, 3)  # batch_size=2, num_samples=16, channels=3

# Case 2: With slices - more complex reconstruction
batch_size = 2
groups = 4
group_samples = 4
channels = 3
tensor_with_slices = torch.randn(batch_size, groups, group_samples, channels)

# Create slices - these define where each block goes in the image
# For a 4x4 image (16 total pixels), we can have 4 slices of 4 pixels each
slices = [
    np.s_[0:2, 0:2],   # top-left 2x2
    np.s_[0:2, 2:4],   # top-right 2x2
    np.s_[2:4, 0:2],   # bottom-left 2x2
    np.s_[2:4, 2:4],   # bottom-right 2x2
]

# Test both scenarios with different parameter combinations
ARGS = {
    "tensor": tensor_no_slices,
    "image_resolution": None,
    "slices": None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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, 4) — shape (2, 3, 4, 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
LocalGlobalBlock Not yet run
unverified (harvested, no recorded run)
maorash/local-global-inrs/modules.py
code served (permissive licence) · get_code("89328da0bc272173")
How it ran

no invocation recorded.

LocallyConnectedBlock Not yet run
unverified (harvested, no recorded run)
maorash/local-global-inrs/modules.py
code served (permissive licence) · get_code("4da48b4400a0bc7a")
How it ran

no invocation recorded.

first_layer_sine_init Not yet run
unverified (harvested, no recorded run)
maorash/local-global-inrs/modules.py
code served (permissive licence) · get_code("58eda4091b7584c9")
How it ran

no invocation recorded.

init_weights_elu Not yet run
unverified (harvested, no recorded run)
maorash/local-global-inrs/modules.py
code served (permissive licence) · get_code("82f21dea7a03c94b")
How it ran

no invocation recorded.

init_weights_normal Not yet run
unverified (harvested, no recorded run)
maorash/local-global-inrs/modules.py
code served (permissive licence) · get_code("cb92d23857fe0a7b")
How it ran

no invocation recorded.

init_weights_selu Not yet run
unverified (harvested, no recorded run)
maorash/local-global-inrs/modules.py
code served (permissive licence) · get_code("25a368af1f6243fe")
How it ran

no invocation recorded.

init_weights_xavier Not yet run
unverified (harvested, no recorded run)
maorash/local-global-inrs/modules.py
code served (permissive licence) · get_code("c5659f39de5ee097")
How it ran

no invocation recorded.

lin2vid Not yet run
unverified (harvested, no recorded run)
maorash/Local-Global-INRs/dataio.py
code served (permissive licence) · get_code("1e72ffd7799784e1")
How it ran

no invocation recorded.

sine_init Not yet run
unverified (harvested, no recorded run)
maorash/local-global-inrs/modules.py
code served (permissive licence) · get_code("34931e02dde165f1")
How it ran

no invocation recorded.

Also present in other papers' code

Of 22 samples lifted from this paper's linked repositories, 22 carry a function identity and 0 do not. Of the 22 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 0 later) and 18 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
laplace 6 other papers
6 dated earlier
2006.09661 (2020-06)
Implicit Neural Representations with Periodic Activation Functions
2007.09601 (2020-07)
Hypersolvers: Toward Fast Continuous-Depth Models
2108.08478 (2021-08)
Learning Anchored Unsigned Distance Functions with Gradient Direction Alignment for Single-view Garment Reconstruction
and 3 more not named here
divergence 4 other papers
4 dated earlier
2006.09661 (2020-06)
Implicit Neural Representations with Periodic Activation Functions
2108.08478 (2021-08)
Learning Anchored Unsigned Distance Functions with Gradient Direction Alignment for Single-view Garment Reconstruction
2303.14092 (2023-03)
NeuFace: Realistic 3D Neural Face Rendering from Multi-view Images
and 1 more not named here
get_mgrid 3 other papers
3 dated earlier
2006.09661 (2020-06)
Implicit Neural Representations with Periodic Activation Functions
2105.02788 (2021-05)
ACORN: Adaptive Coordinate Networks for Neural Scene Representation
2202.10847 (2022-02)
UncertaINR: Uncertainty Quantification of End-to-End Implicit Neural Representations for Computed Tomography
gradient 2 other papers
2 dated earlier
2111.06389 (2021-11)
Full-Body Visual Self-Modeling of Robot Morphologies
2407.09679 (2024-07)
Physics-Informed Learning of Characteristic Trajectories for Smoke Reconstruction

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

Implicit Neural Representations (INRs) have peaked interest in recent years due to their ability to encode natural signals using neural networks. While INRs allow for useful applications such as interpolating new coordinates and signal compression, their black-box nature makes it difficult to modify them post-training. In this paper we explore the idea of editable INRs, and specifically focus on the widely used cropping operation. To this end, we present Local-Global SIRENs -a novel INR architecture that supports cropping by design. Local-Global SIRENs are based on combining local and global feature extraction for signal encoding. What makes their design unique is the ability to effortlessly remove specific portions of an encoded signal, with a proportional weight decrease. This is achieved by eliminating the corresponding weights from the network, without the need for retraining. We further show how this architecture can be used to support the straightforward extension of previously encoded signals. Beyond signal editing, we examine how the Local-Global approach can accelerate training, enhance encoding of various signals, improve downstream performance, and be applied to modern INRs such as INCODE, highlighting its potential and flexibility. Code is available at https://github.com/maorash/Local-Global-INRs.

For agents

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

get_harvested_code_for_paper("2409.19472")
get_code_for_paper("2409.19472")
have("2409.19472")

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

curl https://syntology.ai/api/ran/2409.19472.json

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

[![Syntology run record](https://syntology.ai/api/ran/2409.19472.svg)](https://syntology.ai/paper/2409.19472)

Connect an agent — have() is free.