SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2103.16083 · CVPR · 2021

Fully Convolutional Scene Graph Generation

Ning Yan, Masood Mortazavi, Bir Bhanu, Hengyue Liu

arXiv · PDF · Open in the Atlas

Code that ran

12 of 12 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 11 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2103.16083.json · badge: /api/ran/2103.16083.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
liuhengyue/fcsgg canonical (classifier label)
classifier basis: census
12 ran · 0 no recorded run
FunctionStatusWhere it lives
add_coords Ran
ran; our contract draft was wrong, not the code
liuhengyue/fcsgg/fcsgg/layers/wrappers.py
code served (permissive licence) · get_code("d3c893809080ac97")
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, 5, 4, 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

torch.manual_seed(0)

# Construct a small input tensor: (batch=2, channels=3, x_dim=4, y_dim=5)
input_tensor = torch.randn(2, 3, 4, 5)

ARGS = {
    "input_tensor": input_tensor,
    "with_r": 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 Tensor, ndim 4, dtype family float, shape (2, 5, 4, 5) — shape (2, 5, 4, 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
bbox_iou Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
liuhengyue/fcsgg/fcsgg/evaluation/sgg_evaluation.py
code served (permissive licence) · get_code("b692844d22d4900d")
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 2, dtype family float, shape (2, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create sample bounding boxes in format (x1, y1, x2, y2)
boxes1 = np.array([
    [10, 20, 50, 80],
    [100, 100, 200, 200]
], dtype=np.float32)

boxes2 = np.array([
    [30, 40, 70, 90],
    [110, 110, 210, 210],
    [5, 5, 15, 15]
], dtype=np.float32)

to_move = 1

ARGS = {
    "boxes1": boxes1,
    "boxes2": boxes2,
    "to_move": to_move
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (2, 3) — shape (2, 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
bbox_overlaps Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
liuhengyue/fcsgg/fcsgg/evaluation/sgg_evaluation.py
code served (permissive licence) · get_code("89aaff16a975e560")
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 2, dtype family float, shape (2, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create sample bounding boxes
# boxes1: 2 bounding boxes in format (x1, y1, x2, y2)
boxes1 = np.array([
    [10, 20, 50, 60],
    [30, 40, 80, 90]
], dtype=np.float32)

# boxes2: 3 bounding boxes in format (x1, y1, x2, y2)
boxes2 = np.array([
    [15, 25, 55, 65],
    [100, 100, 200, 200],
    [35, 45, 85, 95]
], dtype=np.float32)

# to_move: small integer parameter used in clipping
to_move = 1

ARGS = {
    "boxes1": boxes1,
    "boxes2": boxes2,
    "to_move": to_move
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (2, 3) — shape (2, 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
conv3x3 Ran
ran; our contract draft was wrong, not the code
liuhengyue/fcsgg/fcsgg/modeling/backbone/dla.py
code served (permissive licence) · get_code("dd1114865f06f0fd")
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 Conv2d. 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

torch.manual_seed(0)

ARGS = {
    "in_planes": 3,
    "out_planes": 64,
    "stride": 1,
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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
corr Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
liuhengyue/fcsgg/fcsgg/layers/rnn.py
code served (permissive licence) · get_code("ea2cbc482c8275e2")
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 6, dtype family float, shape (2, 3, 5, 1, 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

torch.manual_seed(0)

# Create two feature maps with shape (batch, dim, ht, wd)
# batch=2, dim=4, ht=3, wd=5 (distinct sizes for different dimensions)
fmap1 = torch.randn(2, 4, 3, 5)
fmap2 = torch.randn(2, 4, 3, 5)

ARGS = {
    "fmap1": fmap1,
    "fmap2": fmap2,
}
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 6, dtype family float, shape (2, 3, 5, 1, 3, 5) — shape (2, 3, 5, 1, 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
get_model_url Ran
ran; our contract draft was wrong, not the code
liuhengyue/fcsgg/fcsgg/modeling/backbone/dla.py
code served (permissive licence) · get_code("9d627c4b862db734")
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 str. 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 os.path import join

torch.manual_seed(0)

ARGS = {
    'data': 'imagenet',
    'name': 'dla34',
    'hash': 'ba72cf86'
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
intersect_2d Ran
ran, violated the contract we drafted
liuhengyue/fcsgg/fcsgg/evaluation/sgg_evaluation.py
code served (permissive licence) · get_code("6727d2d8f1d79998")
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 2, dtype family float, shape (2, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Construct two 2D numpy arrays with matching number of columns
x1 = np.array([
    [1, 2, 3, 4],
    [5, 6, 7, 8],
], dtype=np.int32)

x2 = np.array([
    [1, 2, 3, 4],
    [9, 10, 11, 12],
    [5, 6, 7, 8],
], dtype=np.int32)

ARGS = {
    'x1': x1,
    'x2': x2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (2, 3) — shape (2, 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
logsumexp_2d Ran
ran; our contract draft was wrong, not the code
liuhengyue/fcsgg/fcsgg/layers/attn.py
code served (permissive licence) · get_code("7abfa32a2f1d7424")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned Tensor, ndim 3, dtype family float, shape (2, 3, 1) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function expects a 3D or higher dimensional tensor
# It accesses tensor.size(0), tensor.size(1), and flattens the rest
# Let's create a 3D tensor with shape (batch=2, dim1=3, dim2=4)
tensor = torch.randn(2, 3, 4)

ARGS = {
    "tensor": tensor
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 3, 1) — shape (2, 3, 1) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
reg_l1_loss Ran
ran; our contract draft was wrong, not the code
liuhengyue/fcsgg/fcsgg/utils/centernet_utils.py
code served (permissive licence) · get_code("0509f439b80d6c9d")
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 0, dtype family float, 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
import torch.nn.functional as F

torch.manual_seed(0)

# output: shape (batch, channel, height, width)
# The function reshapes it to (batch, channel, -1) then permutes to (batch, -1, channel)
# Then indexes with index[0] (batch indices) and index[1] (1d pixel locations)
batch = 2
channel = 4
height = 3
width = 3
output = torch.randn(batch, channel, height, width)

# index: shape (2, N) where N is number of selected pixels
# index[0] contains batch indices, index[1] contains 1d pixel locations
N = 5
batch_indices = torch.tensor([0, 0, 1, 1, 0])
pixel_locations = torch.tensor([2, 5, 1, 7, 4])
index = torch.stack([batch_indices, pixel_locations])

# target: shape (N, 2)
# Should have same second dimension as channel and first dimension matching N
target = torch.randn(N, channel)

ARGS = {
    "output": output,
    "index": index,
    "target": target,
}
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 0, dtype family float, 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
reg_log_l1_loss Ran
ran; our contract draft was wrong, not the code
liuhengyue/fcsgg/fcsgg/utils/centernet_utils.py
code served (permissive licence) · get_code("c9a9bcd3640bffc2")
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 0, dtype family float, 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
import torch.nn.functional as F

torch.manual_seed(0)

# Create a small output tensor: (batch=2, channel=3, height=4, width=5)
output = torch.randn(2, 3, 4, 5, dtype=torch.float32)

# Create index tensors for indexing
# index[0] should be batch indices, index[1] should be spatial indices
# After view and permute, output shape is (batch=2, spatial=20, channel=3)
# We need to select some elements, so create valid indices
batch_indices = torch.tensor([0, 1, 0, 1], dtype=torch.long)
spatial_indices = torch.tensor([0, 5, 10, 15], dtype=torch.long)
index = (batch_indices, spatial_indices)

# Create target tensor matching the selected output shape
# After indexing output[index[0], index[1], :], we get shape (4, 3)
target = torch.ones(4, 3, dtype=torch.float32)

ARGS = {
    "output": output,
    "index": index,
    "target": target,
}
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 0, dtype family float, 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
reg_smooth_l1_loss Ran
ran; our contract draft was wrong, not the code
liuhengyue/fcsgg/fcsgg/utils/centernet_utils.py
code served (permissive licence) · get_code("3773b69c641e1dec")
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 0, dtype family float, 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
import torch.nn.functional as F

torch.manual_seed(0)

# Construct output tensor: (batch=2, channel=4, spatial_dims...)
# Using shape (2, 4, 3, 3) - batch=2, channel=4, spatial=3x3
output = torch.randn(2, 4, 3, 3)

# Construct index tuple for advanced indexing
# index[0] - batch indices, index[2] - spatial indices
# We need to select specific elements from the reshaped tensor
# After view and permute, output becomes (2, 9, 4) where 9 = 3*3 spatial flattened
# index[0] selects batch, index[2] selects spatial position
batch_indices = torch.tensor([0, 1, 0, 1])  # 4 elements
spatial_indices = torch.tensor([0, 1, 2, 3])  # 4 elements
index = (batch_indices, None, spatial_indices)  # index[0] and index[2]

# Construct target tensor
# After indexing output[index[0], index[2], :], shape is (4, 4)
# where 4 is number of selected elements, and 4 is the channel dimension
target = torch.randn(4, 4)

ARGS = {
    "output": output,
    "index": index,
    "target": target,
}
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 0, dtype family float, 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
scenegraph_to_json Ran
ran; our contract draft was wrong, not the code
liuhengyue/fcsgg/fcsgg/evaluation/vg_evaluation.py
code served (permissive licence) · get_code("994959b81c522b99")
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 numpy as np

torch.manual_seed(0)

class Boxes:
    def __init__(self, tensor):
        self.tensor = tensor

class SceneGraph:
    def __init__(self, pred_boxes, scores, pred_classes, rel_scores, rel_inds):
        self.pred_boxes = pred_boxes
        self.scores = scores
        self.pred_classes = pred_classes
        self.rel_scores = rel_scores
        self.rel_inds = rel_inds

# Create minimal tensors with distinct dimensions
pred_boxes_tensor = torch.randn(2, 4)  # 2 boxes, 4 coordinates
boxes = Boxes(pred_boxes_tensor)

scores = torch.tensor([0.9, 0.7])  # 2 scores
pred_classes = torch.tensor([1, 3])  # 2 class ids
rel_scores = torch.tensor([[0.8, 0.6], [0.5, 0.9]])  # 2x2 relationship scores
rel_inds = torch.tensor([[0, 1], [1, 0]])  # 2x2 relationship indices

sg = SceneGraph(
    pred_boxes=boxes,
    scores=scores,
    pred_classes=pred_classes,
    rel_scores=rel_scores,
    rel_inds=rel_inds
)

ARGS = {
    'sg': sg
}
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 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

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 1 the same month, 3 later) and 8 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
conv3x3 686 other papers
274 dated earlier, 10 the same month, 396 later, 6 same year, order not determinable · 31 only as an assembly brick
1406.2661 (2014-06)
Generative Adversarial Networks
1409.0575 (2014-09)
ImageNet Large Scale Visual Recognition Challenge
1412.6572 (2014-12)
Explaining and Harnessing Adversarial Examples
and 683 more not named here
logsumexp_2d 13 other papers
3 dated earlier, 9 later, 1 same year, order not determinable
1806.07366 (2018-06)
Neural Ordinary Differential Equations
1807.06521 (2018-07)
CBAM: Convolutional Block Attention Module
1810.12348 (2018-10)
Gather-Excite: Exploiting Feature Context in Convolutional Neural Networks
and 10 more not named here
intersect_2d 5 other papers
1 dated earlier, 4 later
2004.00436 (2020-04)
Exploring Long Tail Visual Relationship Recognition with Large Vocabulary
2403.17709 (2024-03)
Groupwise Query Specialization and Quality-Aware Multi-Assignment for Transformer-based Visual Relationship Detection
2404.02072 (2024-04)
EGTR: Extracting Graph from Transformer for Scene Graph Generation
and 2 more not named here
get_model_url 1 other paper
1 dated earlier
1904.07850 (2019-04)
Objects as Points

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

for scene graph generation. Comprehensive experiments on the Visual Genome dataset demonstrate the efficacy, efficiency, and generalizability of the proposed method. FC-SGG achieves highly competitive results on recall and zeroshot recall with significantly reduced inference time.

For agents

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

get_harvested_code_for_paper("2103.16083")
get_code_for_paper("2103.16083")
have("2103.16083")

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

curl https://syntology.ai/api/ran/2103.16083.json

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

[![Syntology run record](https://syntology.ai/api/ran/2103.16083.svg)](https://syntology.ai/paper/2103.16083)

Connect an agent — have() is free.