SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.17850 · 2024

FlexiEdit: Frequency-Aware Latent Refinement for Enhanced Non-Rigid Editing

arXiv · PDF · Open in the Atlas

Code that ran

13 of 20 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, 0 violated it and 0 have no contract verdict; 11 failures of our instrument; 7 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/2407.17850.json · badge: /api/ran/2407.17850.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
kookie12/FlexiEdit canonical (classifier label)
classifier basis: census
13 ran · 7 no recorded run
FunctionStatusWhere it lives
calc_mean_std Ran
ran; our contract draft was wrong, not the code
kookie12/FlexiEdit/flexiedit/adain.py
code served (permissive licence) · get_code("7ee9dfbb0bd00480")
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
import torch.nn as nn

torch.manual_seed(0)

# Test case 1: 3D tensor without mask
feat_3d = torch.randn(4, 3, 5)  # C=4, H=3, W=5

# Test case 2: 3D tensor with mask
mask_3d = torch.ones(3, 5)
mask_3d[0, 0] = 0  # Mark one position as masked out

# Test case 3: 2D tensor without mask
feat_2d = torch.randn(4, 15)  # C=4, flattened spatial

# Test case 4: 2D tensor with mask
mask_2d = torch.ones(15)
mask_2d[0] = 0

# Use the 3D case as primary test
ARGS = {
    "feat": feat_3d,
    "eps": 1e-05,
    "mask": 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 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
calc_mean_std_2d Ran
ran; our contract draft was wrong, not the code
kookie12/FlexiEdit/flexiedit/adain.py
code served (permissive licence) · get_code("dcbb6d9b7d233a4e")
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

torch.manual_seed(0)

# Create a 2D feature tensor with shape (C, N) where C=4 channels, N=6 elements
feat = torch.randn(4, 6)

# Create a mask tensor with shape (6,) where 1 indicates valid elements
mask = torch.tensor([1, 1, 1, 0, 1, 1])

ARGS = {
    "feat": feat,
    "eps": 1e-05,
    "mask": mask,
}
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
dilate Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kookie12/FlexiEdit/flexiedit/_ptp_utils.py
code served (permissive licence) · get_code("4f99467fc7580ff1")
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, 1, 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
import torch.nn.functional as F

torch.manual_seed(0)

# Create a small binary image tensor (batch_size=2, channels=1, height=4, width=3)
image = torch.randint(0, 2, (2, 1, 4, 3)).float()

# Define arguments for the dilate function
ARGS = {
    "image": image,
    "kernel_size": 3,
    "stride": 1,
    "padding": 0
}
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, 1, 2, 1) — shape (2, 1, 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
freq_1d Ran
ran; our contract draft was wrong, not the code
kookie12/FlexiEdit/flexiedit/frequency_utils.py
code served (permissive licence) · get_code("158466fd5ab7f865")
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
import torch.fft as fft

torch.manual_seed(0)

# x: diffused latent - should be a tensor
# The function uses fftn with dim=-3, so we need at least 3 dimensions
# Let's create a tensor with shape (batch=2, channels=4, spatial_dim=8)
x = torch.randn(2, 4, 8)

# LPF: low pass filter - should be a tensor with compatible shape for multiplication
# It's multiplied with x_freq, so it should have compatible broadcastable shape
# Since operations happen on dim=-3 (the first dimension in our case after fftn),
# we'll create a filter with the same shape as x for element-wise multiplication
LPF = torch.ones(2, 4, 8) * 0.5  # values between 0 and 1 for a filter

ARGS = {
    "x": x,
    "LPF": LPF,
}
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
freq_2d Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kookie12/FlexiEdit/flexiedit/frequency_utils.py
code served (permissive licence) · get_code("29233e88580c636b")
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.fft as fft

torch.manual_seed(0)

# x: diffused latent - needs to support fftn with dim=(-3, -2, -1)
# So it should be at least 3D. Let's make it 4D: (batch, channels, height, width)
x = torch.randn(2, 4, 3, 5)

# LPF: low pass filter - used in element-wise multiplication with x_freq
# Should have same shape as x_freq (which comes from fftn(x))
LPF = torch.randn(2, 4, 3, 5)

# alpha: used in element-wise multiplication with frequency components
# Should be broadcastable with x_freq
alpha = torch.randn(2, 4, 3, 5)

ARGS = {
    "x": x,
    "LPF": LPF,
    "alpha": alpha,
}
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
freq_3d Ran
ran; our contract draft was wrong, not the code
kookie12/FlexiEdit/flexiedit/frequency_utils.py
code served (permissive licence) · get_code("64614b77d4ec5cec")
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.fft as fft

torch.manual_seed(0)

# Create a small 3D tensor (batch=2, depth=4, height=3, width=5)
x = torch.randn(2, 4, 3, 5)

# Create a low pass filter with same spatial dimensions as x
# Shape: (1, 4, 3, 5) to be broadcastable
LPF = torch.ones(1, 4, 3, 5) * 0.5

ARGS = {
    "x": x,
    "LPF": LPF,
}
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
get_word_inds Ran
ran, honoured the contract we drafted
kookie12/FlexiEdit/inference_flexiedit.py
code served (permissive licence) · get_code("ef18fb78d8588c1d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 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

class MockTokenizer:
    """Mock tokenizer that implements encode and decode methods."""
    def __init__(self):
        # Simple vocabulary mapping
        self.vocab = {
            'hello': 1,
            'world': 2,
            'this': 3,
            'is': 4,
            'a': 5,
            'test': 6,
        }
        self.id_to_token = {v: k for k, v in self.vocab.items()}
        self.id_to_token[0] = '[CLS]'
        self.id_to_token[7] = '[SEP]'
    
    def encode(self, text):
        """Encode text to token IDs. Returns list with [CLS] at start and [SEP] at end."""
        tokens = [0]  # [CLS] token
        words = text.lower().split(' ')
        for word in words:
            # Simple tokenization - map word to ID or use hash
            token_id = self.vocab.get(word, hash(word) % 100 + 10)
            tokens.append(token_id)
        tokens.append(7)  # [SEP] token
        return tokens
    
    def decode(self, token_ids):
        """Decode token IDs back to text."""
        result = []
        for token_id in token_ids:
            if token_id in self.id_to_token:
                result.append(self.id_to_token[token_id])
            else:
                result.append(f"token_{token_id}")
        return ' '.join(result)


# Create setup arguments
text = "hello world this is a test"
word_place_int = 1  # Get indices for word at position 1
tokenizer = MockTokenizer()

ARGS = {
    "text": text,
    "word_place": word_place_int,
    "tokenizer": tokenizer,
}
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 list, ndim 1, dtype family int — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_image Ran
ran, honoured the contract we drafted
kookie12/FlexiEdit/inference_flexiedit.py
code served (permissive licence) · get_code("83dfe7de8ca77508")
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 (1, 3, 512, 512) 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 tempfile
from PIL import Image
import os

torch.manual_seed(0)

# Create a temporary image file for testing
temp_dir = tempfile.mkdtemp()
image_path = os.path.join(temp_dir, "test_image.png")

# Create a simple RGB image (3 channels, 256x256 pixels)
img = Image.new('RGB', (256, 256), color=(100, 150, 200))
img.save(image_path)

device = 'cpu'

ARGS = {
    'image_path': image_path,
    'device': device
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, torch, torchvision — 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 (1, 3, 512, 512) — shape (1, 3, 512, 512) 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
make_grid Ran
ran; our contract draft was wrong, not the code
kookie12/FlexiEdit/flexiedit/utils.py
code served (permissive licence) · get_code("68eb651d672fa27f")
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 (3, 6, 13) 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)

# Create a simple 4D tensor: (B x C x H x W)
# B=2 (batch size), C=3 (channels), H=4 (height), W=5 (width)
tensor = torch.randn(2, 3, 4, 5)

# Create alternative test cases
tensor_list = [torch.randn(3, 4, 5), torch.randn(3, 4, 5)]
tensor_2d = torch.randn(4, 5)
tensor_3d = torch.randn(1, 4, 5)
tensor_single_channel = torch.randn(2, 1, 4, 5)

ARGS = {
    "tensor": tensor,
    "nrow": 2,
    "padding": 1,
    "normalize": False,
    "value_range": None,
    "scale_each": False,
    "pad_value": 0.0,
}
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 (3, 6, 13) — shape (3, 6, 13) 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
masked_adain Ran
ran; our contract draft was wrong, not the code
kookie12/FlexiEdit/flexiedit/adain.py
code served (permissive licence) · get_code("cb4f657896b3cfe3")
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 (4, 3, 5) 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 feature tensors
# The function expects style_feat and content_feat to have matching first 2 dimensions
# They can be either 2D (C, N) or 3D (C, H, W)
# Let's create 3D tensors: (channels, height, width)

C = 4  # channels
H = 3  # height
W = 5  # width

style_feat = torch.randn(C, H, W)
content_feat = torch.randn(C, H, W)

# Create masks - binary masks same spatial dims as features
# mask values should be 0 or 1
style_mask = torch.randint(0, 2, (H, W)).float()
content_mask = torch.randint(0, 2, (H, W)).float()

ARGS = {
    "style_feat": style_feat,
    "content_feat": content_feat,
    "style_mask": style_mask,
    "content_mask": content_mask,
}
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 (4, 3, 5) — shape (4, 3, 5) 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
slerp Ran
ran; our contract draft was wrong, not the code
kookie12/FlexiEdit/flexiedit/_ptp_utils.py
code served (permissive licence) · get_code("2eb4b43093b866ae")
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 with batch size 2 and dimension 4
# val: interpolation parameter (scalar between 0 and 1)
val = 0.5

# low: batch of 2 vectors, each of dimension 4
low = torch.randn(2, 4)

# high: batch of 2 vectors, each of dimension 4
high = torch.randn(2, 4)

ARGS = {
    "val": val,
    "low": low,
    "high": high,
}
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
slerp_tensor Ran
ran; our contract draft was wrong, not the code
kookie12/FlexiEdit/flexiedit/_ptp_utils.py
code served (permissive licence) · get_code("eabfbcd7d2863c2b")
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)

# slerp_tensor takes three parameters:
# - val: a scalar interpolation value
# - low: a tensor (will be flattened to 2D in the function)
# - high: a tensor (will be flattened to 2D in the function)

# Create small tensors with distinct shapes
# Batch size 2, and we'll use a multi-dimensional shape to test reshaping
val = 0.5
low = torch.randn(2, 4)
high = torch.randn(2, 4)

ARGS = {
    "val": val,
    "low": low,
    "high": high,
}
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
tensor2numpy Ran
ran; our contract draft was wrong, not the code
kookie12/FlexiEdit/flexiedit/utils.py
code served (permissive licence) · get_code("94c181218397b5c3")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 3, dtype family int, shape (4, 10, 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 batch of image tensors
# Shape: (B=2, C=3, H=4, W=5) - 2 images, 3 channels (RGB), 4x5 pixels
tensor = torch.rand(2, 3, 4, 5)

# Define ARGS dict with all parameters for tensor2numpy
ARGS = {
    "tensor": tensor,
    "format": None,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (4, 10, 3) — shape (4, 10, 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
add_text_to_image Not yet run
unverified (harvested, no recorded run)
kookie12/FlexiEdit/flexiedit/utils.py
code served (permissive licence) · get_code("1af043b99331d6f5")
How it ran

no invocation recorded.

expand_adj_noun_pairs Not yet run
unverified (harvested, no recorded run)
kookie12/FlexiEdit/flexiedit/get_edited_words.py
code served (permissive licence) · get_code("540f946754cd6ce8")
How it ran

no invocation recorded.

get_edited_phrases Not yet run
unverified (harvested, no recorded run)
kookie12/FlexiEdit/flexiedit/get_edited_words.py
code served (permissive licence) · get_code("6e0f6f56c7466627")
How it ran

no invocation recorded.

register_conv_control_efficient Not yet run
unverified (harvested, no recorded run)
kookie12/FlexiEdit/flexiedit/attention_register.py
code served (permissive licence) · get_code("e97c3effd6355f80")
How it ran

no invocation recorded.

regiter_attention_editor_diffusers Not yet run
unverified (harvested, no recorded run)
kookie12/FlexiEdit/flexiedit/attention_register.py
code served (permissive licence) · get_code("827413ab1b8cdef0")
How it ran

no invocation recorded.

regiter_attention_editor_ldm Not yet run
unverified (harvested, no recorded run)
kookie12/FlexiEdit/flexiedit/attention_register.py
code served (permissive licence) · get_code("431f768e76ae86fa")
How it ran

no invocation recorded.

remove_unnecessary_words Not yet run
unverified (harvested, no recorded run)
kookie12/FlexiEdit/flexiedit/get_edited_words.py
code served (permissive licence) · get_code("24cb77e71d1f327e")
How it ran

no invocation recorded.

Also present in other papers' code

Of 20 samples lifted from this paper's linked repositories, 20 carry a function identity and 0 do not. Of the 20 distinct functions with an identity, 9 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 0 the same month, 4 later) and 11 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
regiter_attention_editor_ldm 4 other papers
3 dated earlier, 1 later
2304.08465 (2023-04)
MasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editing
2311.16491 (2023-11)
$Z^*$: Zero-shot Style Transfer via Attention Rearrangement
2312.02212 (2023-12)
Portrait Diffusion: Training-free Face Stylization with Chain-of-Painting
and 1 more not named here
calc_mean_std 2 other papers
1 dated earlier, 1 later
2401.02126 (2024-01)
Unified Diffusion-Based Rigid and Non-Rigid Editing with Text and Image Guidance
2602.03625 (2026-02)
Multi-Objective Optimization for Synthetic-to-Real Style Transfer
calc_mean_std_2d 2 other papers
1 dated earlier, 1 later
2401.02126 (2024-01)
Unified Diffusion-Based Rigid and Non-Rigid Editing with Text and Image Guidance
2602.03625 (2026-02)
Multi-Objective Optimization for Synthetic-to-Real Style Transfer
slerp 2 other papers
2 dated earlier
1912.04958 (2019-12)
Analyzing and Improving the Image Quality of StyleGAN
2403.11105 (2024-03)
Source Prompt Disentangled Inversion for Boosting Image Editability with Diffusion Models
freq_2d 1 other paper
1 later
2501.11319 (2025-01)
StyleSSP: Sampling StartPoint Enhancement for Training-free Diffusion-based Method for Style Transfer
masked_adain 1 other paper
1 dated earlier
2401.02126 (2024-01)
Unified Diffusion-Based Rigid and Non-Rigid Editing with Text and Image Guidance
register_conv_control_efficient 1 other paper
1 dated earlier
2401.02126 (2024-01)
Unified Diffusion-Based Rigid and Non-Rigid Editing with Text and Image Guidance
regiter_attention_editor_diffusers 1 other paper
1 dated earlier
2401.02126 (2024-01)
Unified Diffusion-Based Rigid and Non-Rigid Editing with Text and Image Guidance
slerp_tensor 1 other paper
1 dated earlier
2403.11105 (2024-03)
Source Prompt Disentangled Inversion for Boosting Image Editability with Diffusion Models

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

Current image editing methods primarily utilize DDIM Inversion, employing a two-branch diffusion approach to preserve the attributes and layout of the original image. However, these methods encounter challenges with non-rigid edits, which involve altering the image's layout or structure. Our comprehensive analysis reveals that the high-frequency components of DDIM latent, crucial for retaining the original image's key features and layout, significantly contribute to these limitations. Addressing this, we introduce FlexiEdit, which enhances fidelity to input text prompts by refining DDIM latent, by reducing high-frequency components in targeted editing areas. FlexiEdit comprises two key components: (1) Latent Refinement, which modifies DDIM latent to better accommodate layout adjustments, and (2) Edit Fidelity Enhancement via Re-inversion, aimed at ensuring the edits more accurately reflect the input text prompts. Our approach represents notable progress in image editing, particularly in performing complex non-rigid edits, showcasing its enhanced capability through comparative experiments.

For agents

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

get_harvested_code_for_paper("2407.17850")
get_code_for_paper("2407.17850")
have("2407.17850")

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

curl https://syntology.ai/api/ran/2407.17850.json

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

[![Syntology run record](https://syntology.ai/api/ran/2407.17850.svg)](https://syntology.ai/paper/2407.17850)

Connect an agent — have() is free.