SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2103.17202 · CVPR · 2021

GrooMeD-NMS: Grouped Mathematically Differentiable NMS for Monocular 3D Object Detection

Xiaoming Liu, Abhinav Kumar, Garrick Brazil

arXiv · PDF · Open in the Atlas

Code that ran

5 of 7 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; 4 failures of our instrument; 2 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/2103.17202.json · badge: /api/ran/2103.17202.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
abhi1kumar/groomed_nms marked official in Papers with Code
Papers with Code archive
5 ran · 2 no recorded run
FunctionStatusWhere it lives
cast_to_cpu_cuda_tensor Ran
ran; our contract draft was wrong, not the code
abhi1kumar/groomed_nms/lib/groomed_nms.py
code served (permissive licence) · get_code("04ad3cb3cbd977c2")
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. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, 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 small CPU tensors
input_tensor = torch.randn(2, 4)
reference_tensor = torch.randn(2, 4)

ARGS = {
    "input": input_tensor,
    "reference_tensor": reference_tensor,
}
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
get_groups Ran
ran; our contract draft was wrong, not the code
abhi1kumar/groomed_nms/lib/groomed_nms.py
code served (permissive licence) · get_code("43c8ee53afde7f84")
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. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create small tensors for testing
num_boxes = 5

# Create a small IoU matrix (symmetric, values between 0 and 1)
iou_unsorted = torch.tensor([
    [1.0, 0.8, 0.3, 0.2, 0.1],
    [0.8, 1.0, 0.4, 0.15, 0.2],
    [0.3, 0.4, 1.0, 0.6, 0.5],
    [0.2, 0.15, 0.6, 1.0, 0.7],
    [0.1, 0.2, 0.5, 0.7, 1.0]
], dtype=torch.float32)

# Create scores for the boxes
scores_unsorted = torch.tensor([0.9, 0.8, 0.7, 0.6, 0.5], dtype=torch.float32)

# Group threshold for IoU comparison
group_threshold = 0.5

# Maximum group size
group_size = 100

# Whether to return original indices
return_original_indices = True

ARGS = {
    "iou_unsorted": iou_unsorted,
    "group_threshold": group_threshold,
    "scores_unsorted": scores_unsorted,
    "group_size": group_size,
    "return_original_indices": return_original_indices,
}
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 list — a list output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
pruning_function Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
abhi1kumar/groomed_nms/lib/groomed_nms.py
code served (permissive licence) · get_code("d9ff1aadcc385d0f")
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, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import torch

torch.manual_seed(0)

# Test with torch.Tensor
iou_tensor = torch.rand(2, 3)

# Test with numpy.ndarray
iou_numpy = np.random.rand(2, 3)

# Define ARGS with all parameters
ARGS = {
    'iou': iou_tensor,
    'nms_threshold': 0.4,
    'temperature': 0.01,
    'pruning_method': 'sigmoidal'
}
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 (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
sigmoid_numpy Ran
ran, violated the contract we drafted
abhi1kumar/groomed_nms/lib/groomed_nms.py
code served (permissive licence) · get_code("f8a26ac1c00da1b4")
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 1, dtype family float, shape (8,) 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

def sigmoid_numpy(x):
    y = np.zeros(x.shape)
    index_gt = np.where(x > 0)[0]
    y[index_gt] = 1.0 / (1 + np.exp(-x[index_gt]))
    index_lt = np.where(x <= 0)[0]
    y[index_lt] = np.exp(x[index_lt]) / (1 + np.exp(x[index_lt]))
    return y

# Create a small numpy array with mixed positive and negative values
x = np.array([-2.0, -0.5, 0.0, 1.5, 2.0, -1.0, 0.5, 3.0], dtype=np.float32)

ARGS = {
    "x": x
}
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 1, dtype family float, shape (8,) — shape (8,) 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
soft_sort Ran
ran; our contract draft was wrong, not the code
abhi1kumar/groomed_nms/lib/groomed_nms.py
code served (permissive licence) · get_code("94bbd14e775148be")
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. 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)

# Create a small scores tensor (N,)
scores = torch.randn(4)

# Create a full_matrix tensor that will be sorted along with scores
# This should have shape (N, M) where N matches scores dimension
full_matrix = torch.randn(4, 3)

# Temperature parameter for soft sorting
temperature = 0.01

ARGS = {
    "scores": scores,
    "full_matrix": full_matrix,
    "temperature": temperature,
}
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 — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
differentiable_nms Not yet run
unverified (harvested, no recorded run)
abhi1kumar/groomed_nms/lib/groomed_nms.py
code served (permissive licence) · get_code("f4865c086d476c14")
How it ran

no invocation recorded.

indices_copy Not yet run
unverified (harvested, no recorded run)
abhi1kumar/groomed_nms/lib/groomed_nms.py
code served (permissive licence) · get_code("a18e3e8e59a26e46")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 7 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.

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

Overview of our method. (a) Conventional object detection has a mismatch between training and inference as it uses NMS only in inference. (b) To address this, we propose a novel GrooMeD-NMS layer, such that the network is trained end-to-end with NMS applied. s and r denote the score of boxes B before and after the NMS respectively. O denotes the matrix containing IoU2D overlaps of B. L bef ore denotes the losses before the NMS, while L af ter denotes the loss after the NMS. (c) GrooMeD-NMS layer calculates r in a differentiable manner giving gradients from L af ter when the best-localized box corresponding to an object is not selected after NMS.

For agents

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

get_harvested_code_for_paper("2103.17202")
get_code_for_paper("2103.17202")
have("2103.17202")

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

curl https://syntology.ai/api/ran/2103.17202.json

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

[![Syntology run record](https://syntology.ai/api/ran/2103.17202.svg)](https://syntology.ai/paper/2103.17202)

Connect an agent — have() is free.