SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2405.17211 · ICLR · 2025

Spectral-Refiner: Accurate Fine-Tuning of Spatiotemporal Fourier Neural Operator for Turbulent Flows

Shuhao Cao, Yuanzhe Xi, Rui Li, Francesco Brarda

arXiv · PDF · Open in the Atlas

Code that ran

14 of 15 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; 5 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 8 failures of our instrument; 1 no recorded run), measured 2026-11-22; 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/2405.17211.json · badge: /api/ran/2405.17211.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
scaomath/torch-cfd canonical (classifier label)
classifier basis: census
12 ran · 1 no recorded run
neuraloperator/neuraloperator canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
copy not recorded community
not in the paper's declared repositories
1 ran · 0 no recorded run
FunctionStatusWhere it lives
central_diff Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
scaomath/torch-cfd/fno/losses.py
code served (permissive licence) · get_code("c17bf7dfdb45871b")
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 5, 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
import torch.nn.functional as F

torch.manual_seed(0)

# Create a small 4D tensor (batch=2, channels=1, height=4, width=4)
u = torch.randn(2, 1, 4, 4)

ARGS = {
    'u': u,
    'h': None,
    'mode': 'constant',
    'padding': True,
    'value': None,
    'channel_last': False,
}
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 tuple, ndim 5, dtype family float — 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
default Ran
ran, honoured the contract we drafted
scaomath/torch-cfd/torch_cfd/advection.py
code served (permissive licence) · get_code("5a5ec93138deef29")
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 int, ndim 0, dtype family int, shape () 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)

ARGS = {
    "value": 42,
    "d": 100
}
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 int, ndim 0, dtype family int, shape () — shape () 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
fft_expand_dims Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
scaomath/torch-cfd/torch_cfd/spectral.py
code served (permissive licence) · get_code("f82b9bc3c0e5b177")
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 5, 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
import numpy as np
from einops import repeat

torch.manual_seed(0)

# fft_expand_dims expects:
# - fft_mesh: a tuple of two 2D arrays/tensors (kx, ky)
# - batch_size: an integer

# Create small 2D tensors for kx and ky
kx = torch.randn(4, 3)  # shape (x=4, y=3)
ky = torch.randn(4, 3)  # shape (x=4, y=3)

fft_mesh = (kx, ky)
batch_size = 2

ARGS = {
    "fft_mesh": fft_mesh,
    "batch_size": batch_size,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, 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 5, dtype family float — 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
fft_mesh_2d Ran
ran, honoured the contract we drafted
scaomath/torch-cfd/torch_cfd/spectral.py
code served (permissive licence) · get_code("81d554543fe95eda")
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 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
import torch.fft as fft

torch.manual_seed(0)

ARGS = {
    'n': 8,
    'diam': 2.0,
    'device': None,
}
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 tuple, ndim 3, dtype family float — 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
log_normal_density Ran
ran; our contract draft was wrong, not the code
scaomath/torch-cfd/torch_cfd/initial_conditions.py
code served (permissive licence) · get_code("30dc8b877f782e80")
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 1, dtype family float, shape (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 math

torch.manual_seed(0)

# The function log_normal_density takes:
# - k: a tensor (used with torch.log and torch operations)
# - mode: a float
# - variance: optional float (default 0.25)

# Create a small tensor for k (must be positive for torch.log)
k = torch.tensor([0.5, 1.0, 2.0, 3.0], dtype=torch.float32)

# Create a float for mode
mode = 1.5

# variance has a default of 0.25, but we can specify it
variance = 0.25

ARGS = {
    "k": k,
    "mode": mode,
    "variance": variance,
}
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 1, dtype family float, shape (4,) — shape (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
outer_sum Ran
ran, honoured the contract we drafted
scaomath/torch-cfd/torch_cfd/solvers.py
code served (permissive licence) · get_code("f744cc7a9cfb3405")
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
from typing import Sequence

torch.manual_seed(0)

# Create a sequence of small 1D tensors
# outer_sum uses reduce with lambda a, b: a[..., None] + b
# Each tensor will be expanded with None dimensions as we go through the reduction
x = [
    torch.randn(2),      # Shape: (2,)
    torch.randn(3),      # Shape: (3,)
    torch.randn(4),      # Shape: (4,)
]

ARGS = {
    "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 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
partialclass Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from neuraloperator/neuraloperator
pointer only · get_code("fc02f5f0a7b4896e")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

partialclass Ran
ran; our contract draft was wrong, not the code
neuraloperator/neuraloperator/neuralop/models/fno.py
code served (permissive licence) · get_code("2041982f0cf2a6be")
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 type. 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 functools import partialmethod

# Create a simple base class to use with partialclass
class SimpleModel(nn.Module):
    def __init__(self, input_dim=10, hidden_dim=20, output_dim=5):
        super().__init__()
        self.linear1 = nn.Linear(input_dim, hidden_dim)
        self.linear2 = nn.Linear(hidden_dim, output_dim)
    
    def forward(self, x):
        x = torch.relu(self.linear1(x))
        x = self.linear2(x)
        return x

# Set seed for reproducibility
torch.manual_seed(0)

# Define arguments for partialclass
ARGS = {
    'new_name': 'CustomModel',
    'cls': SimpleModel,
    'args': (),
    'kwargs': {
        'input_dim': 8,
        'hidden_dim': 16,
        'output_dim': 4
    }
}
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 type — a type output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a type 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
safe_div Ran
ran; our contract draft was wrong, not the code
scaomath/torch-cfd/torch_cfd/advection.py
code served (permissive licence) · get_code("f23abec4f65ce479")
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

torch.manual_seed(0)

# Create small tensors for the safe_div function
x = torch.tensor([[1.0, 2.0, 3.0, 4.0],
                   [5.0, 6.0, 7.0, 8.0]])

y = torch.tensor([[1.0, 0.0, 2.0, 0.0],
                   [4.0, 2.0, 0.0, 1.0]])

default_numerator = 1

ARGS = {
    "x": x,
    "y": y,
    "default_numerator": default_numerator
}
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
slice_along_axis Ran
ran, honoured the contract we drafted
scaomath/torch-cfd/torch_cfd/tensor_utils.py
code served (permissive licence) · get_code("c0504fbcda5ca095")
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 (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)

# Test case 1: Single tensor with integer index
tensor_1 = torch.randn(2, 4, 3)
idx_1 = 1
axis_1 = 0
expect_same_dims_1 = True

ARGS = {
    "inputs": tensor_1,
    "axis": axis_1,
    "idx": idx_1,
    "expect_same_dims": expect_same_dims_1,
}
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 (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
spectral_laplacian_2d Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
scaomath/torch-cfd/torch_cfd/spectral.py
code served (permissive licence) · get_code("0fb57bced213c0a9")
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, 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
import numpy as np

torch.manual_seed(0)

# Create fft_mesh: a tuple of (kx, ky) tensors
# These represent frequency domain meshes for a 2D FFT
# Shape: (batch, height, width) or similar
kx = torch.randn(2, 4, 3)
ky = torch.randn(2, 4, 3)
fft_mesh = (kx, ky)

device = None

ARGS = {
    "fft_mesh": fft_mesh,
    "device": device,
}
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 3, dtype family float, shape (2, 4, 3) — shape (2, 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
split_along_axis Ran
ran, violated the contract we drafted
scaomath/torch-cfd/torch_cfd/tensor_utils.py
code served (permissive licence) · get_code("cc8238497f7c4a27")
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 4, 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 test tensors with distinct dimensions
tensor1 = torch.randn(2, 4, 3)  # batch=2, dim1=4, dim2=3
tensor2 = torch.randn(2, 4, 5)  # batch=2, dim1=4, dim2=5

# Test case 1: split a single tensor along axis 0
# This tests the basic tensor case

ARGS = {
    "inputs": tensor1,
    "split_idx": 1,
    "axis": 0,
    "expect_same_dims": True,
}
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 tuple, ndim 4, dtype family float — 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
split_axis Ran
ran, honoured the contract we drafted
scaomath/torch-cfd/torch_cfd/tensor_utils.py
code served (permissive licence) · get_code("3976415f0a6bab34")
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 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 collections.abc import Sequence

torch.manual_seed(0)

# Test case 1: single tensor input
tensor_input = torch.randn(2, 4, 3)

# Test case 2: sequence of tensors with consistent dimension
seq_input = [torch.randn(2, 4, 3), torch.randn(2, 4, 3)]

# Test case 3: sequence with tensors and non-tensor items
seq_mixed = [torch.randn(2, 4, 3), "label", torch.randn(2, 4, 3)]

ARGS = {
    "inputs": tensor_input,
    "dim": 1,
    "keep_dims": False,
}
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 tuple, ndim 3, dtype family float — 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
van_leer_limiter Ran
ran; our contract draft was wrong, not the code
scaomath/torch-cfd/torch_cfd/advection.py
code served (permissive licence) · get_code("f9dfe180c45e72ca")
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

torch.manual_seed(0)

# van_leer_limiter takes a single parameter 'r' which is used in:
# - r > 0 comparison
# - safe_div(2 * r, 1 + r) computation
# So r should be a tensor that supports these operations

r = torch.randn(2, 4)

ARGS = {
    "r": r
}
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
scalar_potential Not yet run
unverified (harvested, no recorded run)
scaomath/torch-cfd/torch_cfd/forcings.py
code served (permissive licence) · get_code("57a6d2f9473ef5e3")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 1 later) and 12 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
partialclass 2 other papers
1 dated earlier, 1 later
2307.15034 (2023-07)
Guaranteed Approximation Bounds for Mixed-Precision Neural Operators
2503.17797 (2025-03)
Enhancing Fourier Neural Operators with Local Spatial Features
default 1 other paper
1 dated earlier
1912.04958 (2019-12)
Analyzing and Improving the Image Quality of StyleGAN

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

Recent advancements in operator-type neural networks have shown promising results in approximating the solutions of spatiotemporal Partial Differential Equations (PDEs). However, these neural networks often entail considerable training expenses, and may not always achieve the desired accuracy required in many scientific and engineering disciplines. In this paper, we propose a new learning framework to address these issues. A new spatiotemporal adaptation is proposed to generalize any Fourier Neural Operator (FNO) variant to learn maps between Bochner spaces, which can perform an arbitrary-length temporal super-resolution for the first time. To better exploit this capacity, a new paradigm is proposed to refine the commonly adopted end-to-end neural operator training and evaluations with the help from the wisdom from traditional numerical PDE theory and techniques. Specifically, in the learning problems for the turbulent flow modeled by the Navier-Stokes Equations (NSE), the proposed paradigm trains an FNO only for a few epochs. Then, only the newly proposed spatiotemporal spectral convolution layer is fine-tuned without the frequency truncation. The spectral fine-tuning loss function uses a negative Sobolev norm for the first time in operator learning, defined through a reliable functional-type a posteriori error estimator whose evaluation is exact thanks to the Parseval identity. Moreover, unlike the difficult nonconvex optimization problems in the end-to-end training, this fine-tuning loss is convex. Numerical experiments on commonly used NSE benchmarks demonstrate significant improvements in both computational efficiency and accuracy, compared to end-to-end evaluation and traditional numerical PDE solvers under certain conditions. The source code is publicly available at https://github.com/scaomath/torch-cfd.

For agents

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

get_harvested_code_for_paper("2405.17211")
get_code_for_paper("2405.17211")
have("2405.17211")

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

curl https://syntology.ai/api/ran/2405.17211.json

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

[![Syntology run record](https://syntology.ai/api/ran/2405.17211.svg)](https://syntology.ai/paper/2405.17211)

Connect an agent — have() is free.