SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.03459 · 2024

LW-DETR: A Transformer Replacement to YOLO for Real-Time Detection

arXiv · PDF · Open in the Atlas

Code that ran

13 of 16 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; 3 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/2406.03459.json · badge: /api/ran/2406.03459.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
atten4vis/lw-detr canonical (classifier label)
classifier basis: census
13 ran · 3 no recorded run
FunctionStatusWhere it lives
build_o365 Ran
ran; our contract draft was wrong, not the code
atten4vis/lw-detr/datasets/o365.py
code served (permissive licence) · get_code("53a9978ad1f02a68")
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 CocoDetection. 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
from pathlib import Path
from unittest.mock import MagicMock, patch

torch.manual_seed(0)

# Create mock classes for CocoDetection and transforms
class CocoDetection:
    def __init__(self, img_folder, ann_file, transforms=None):
        self.img_folder = img_folder
        self.ann_file = ann_file
        self.transforms = transforms

def make_coco_transforms(image_set):
    return MagicMock()

def make_coco_transforms_square_div_64(image_set):
    return MagicMock()

# Patch the CocoDetection and transform functions in the module
import sys
current_module = sys.modules[__name__]
current_module.CocoDetection = CocoDetection
current_module.make_coco_transforms = make_coco_transforms
current_module.make_coco_transforms_square_div_64 = make_coco_transforms_square_div_64

# Create args object
class Args:
    def __init__(self):
        self.coco_path = Path('/tmp/coco')
        self.square_resize = False
        self.square_resize_div_64 = False

args = Args()

ARGS = {
    'image_set': 'train',
    'args': args
}
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 CocoDetection — a CocoDetection output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a CocoDetection 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
convert_to_xywh Ran
ran, honoured the contract we drafted
atten4vis/lw-detr/datasets/coco_eval.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from yuhangzang/contextdet, is pointer only (licence: NOASSERTION) · get_code("f31a58bf6457ced5")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from yuhangzang/contextdet, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

convert_to_xywh Ran
ran, honoured the contract we drafted
atten4vis/lw-detr/deploy/benchmark.py
code served (permissive licence) · get_code("0cf40c7d0cb18b12")
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)

boxes = torch.tensor([
    [10.0, 20.0, 50.0, 70.0],
    [5.0, 15.0, 40.0, 60.0]
], dtype=torch.float32)

ARGS = {
    "boxes": boxes
}
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
crop Ran
ran; our contract draft was wrong, not the code
atten4vis/lw-detr/datasets/transforms.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from kellyiss/situformer, is pointer only (licence: NONE) · get_code("74f3f5c1bab38c14")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from kellyiss/situformer, and GitHub detected no licence in that repository when it was checked. 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.

evaluate Ran
ran; our contract draft was wrong, not the code
atten4vis/lw-detr/datasets/coco_eval.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from yuhangzang/contextdet, is pointer only (licence: NOASSERTION) · get_code("fe0ddcc2d420c9a0")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from yuhangzang/contextdet, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

gen_encoder_output_proposals Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
atten4vis/lw-detr/models/transformer.py
code served (permissive licence) · get_code("515cc5f258dc7853")
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)

# Setup: Create minimal valid inputs for gen_encoder_output_proposals
# The function expects:
# - memory: (bs, sum{hw}, d_model)
# - memory_padding_mask: (bs, sum{hw}) or None
# - spatial_shapes: (nlevel, 2) with [H, W] for each level
# - unsigmoid: bool

# Let's create a simple 2-level feature pyramid
# Level 0: 4x3 spatial dims
# Level 1: 2x2 spatial dims
batch_size = 2
d_model = 4

# Spatial shapes: 2 levels with different H, W
spatial_shapes = torch.tensor([[4, 3], [2, 2]], dtype=torch.long)

# Total sequence length: 4*3 + 2*2 = 12 + 4 = 16
total_hw = 4 * 3 + 2 * 2  # 16

# Memory tensor: (batch_size, sum{hw}, d_model)
memory = torch.randn(batch_size, total_hw, d_model)

# Memory padding mask: (batch_size, sum{hw})
# False means valid, True means padding
memory_padding_mask = torch.zeros(batch_size, total_hw, dtype=torch.bool)
# Mark some positions as padding
memory_padding_mask[0, 14:16] = True
memory_padding_mask[1, 15] = True

ARGS = {
    "memory": memory,
    "memory_padding_mask": memory_padding_mask,
    "spatial_shapes": spatial_shapes,
    "unsigmoid": 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
gen_sineembed_for_position Ran
ran; our contract draft was wrong, not the code
atten4vis/lw-detr/models/transformer.py
code served (permissive licence) · get_code("dbaf5088c2722765")
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, 256) 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 small position tensor with shape (batch=2, num_positions=3, coordinates=2)
# The last dimension can be 2 or 4 based on the function logic
pos_tensor_2d = torch.randn(2, 3, 2)

# Test with 2D coordinates (x, y)
ARGS = {
    "pos_tensor": pos_tensor_2d,
    "dim": 128
}
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, 256) — shape (2, 3, 256) 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_coco_pretrain_from_obj365 Ran
ran; our contract draft was wrong, not the code
atten4vis/lw-detr/util/obj365_to_coco_model.py
code served (permissive licence) · get_code("dddb7619ec008b7f")
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 (100, 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)

# The function takes two tensors and performs indexing/assignment operations
# cur_tensor needs to support indexing and assignment
# pretrain_tensor needs to support indexing
# Both need .size() method
# cur_tensor needs .requires_grad attribute

# Create tensors with distinct sizes for probing
# cur_tensor should have at least 80 elements (max coco_id is 90, so needs size > 90)
# pretrain_tensor should have at least 329 elements (max obj_id + 1 is 328 + 1 = 329)

cur_tensor = torch.randn(100, 4)
pretrain_tensor = torch.randn(330, 4)

ARGS = {
    "cur_tensor": cur_tensor,
    "pretrain_tensor": pretrain_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 (100, 4) — shape (100, 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_image_list Ran
ran; our contract draft was wrong, not the code
atten4vis/lw-detr/deploy/benchmark.py
code served (permissive licence) · get_code("aaf8532df82e8e3f")
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 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 json
import tempfile
import os

# Create a temporary JSON file with the expected structure
temp_dir = tempfile.mkdtemp()
ann_file = os.path.join(temp_dir, "annotations.json")

# Create a valid annotation file with images data
annotations_data = {
    "images": [
        {"id": 1, "file_name": "image1.jpg", "width": 640, "height": 480},
        {"id": 2, "file_name": "image2.jpg", "width": 640, "height": 480}
    ]
}

with open(ann_file, 'w') as f:
    json.dump(annotations_data, f)

ARGS = {
    "ann_file": ann_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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 float — 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
hflip Ran
ran; our contract draft was wrong, not the code
atten4vis/lw-detr/datasets/transforms.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from kellyiss/situformer, is pointer only (licence: NONE) · get_code("9be9e658b5bd688a")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from kellyiss/situformer, and GitHub detected no licence in that repository when it was checked. 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.

position_supervised_loss Ran
ran; our contract draft was wrong, not the code
atten4vis/lw-detr/models/lwdetr.py
code served (permissive licence) · get_code("09ea33b5c4fc4241")
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 small tensors for the function
batch_size = 2
num_features = 4

# inputs: logits tensor that will be passed through sigmoid
inputs = torch.randn(batch_size, num_features)

# targets: binary targets (0 or 1)
targets = torch.randint(0, 2, (batch_size, num_features)).float()

# num_boxes: number of boxes for normalization
num_boxes = 2

# alpha: focal loss weighting parameter
alpha = 0.25

# gamma: focal loss exponent
gamma = 2.0

ARGS = {
    'inputs': inputs,
    'targets': targets,
    'num_boxes': num_boxes,
    'alpha': alpha,
    'gamma': gamma,
}
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
sigmoid_focal_loss Ran
ran; our contract draft was wrong, not the code
atten4vis/lw-detr/models/lwdetr.py
code served (permissive licence) · get_code("5c0711aada67957e")
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 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 small tensors for inputs and targets
# inputs: predictions (logits) - shape (2, 4) for batch_size=2, num_predictions=4
inputs = torch.randn(2, 4, requires_grad=True)

# targets: binary labels - shape (2, 4) matching inputs
targets = torch.randint(0, 2, (2, 4), dtype=torch.float32)

# num_boxes: number of boxes for normalization
num_boxes = 2

# alpha and gamma are already float parameters with defaults
alpha = 0.25
gamma = 2.0

ARGS = {
    "inputs": inputs,
    "targets": targets,
    "num_boxes": num_boxes,
    "alpha": alpha,
    "gamma": gamma,
}
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 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
sigmoid_varifocal_loss Ran
ran; our contract draft was wrong, not the code
atten4vis/lw-detr/models/lwdetr.py
code served (permissive licence) · get_code("e1af8817e1aa2d5e")
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 small tensors for the function
batch_size = 2
num_classes = 4

# inputs: logits tensor (batch_size, num_classes)
inputs = torch.randn(batch_size, num_classes)

# targets: binary targets tensor (batch_size, num_classes)
targets = torch.randint(0, 2, (batch_size, num_classes)).float()

# num_boxes: number of boxes for normalization
num_boxes = 2

ARGS = {
    "inputs": inputs,
    "targets": targets,
    "num_boxes": num_boxes,
    "alpha": 0.25,
    "gamma": 2.0
}
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 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
build_o365_raw Not yet run
unverified (harvested, no recorded run)
atten4vis/lw-detr/datasets/o365.py
code served (permissive licence) · get_code("bb1c621e43ea7529")
How it ran

no invocation recorded.

build_transformer Not yet run
unverified (harvested, no recorded run)
atten4vis/lw-detr/models/transformer.py
code served (permissive licence) · get_code("ec7a90cc24d6630d")
How it ran

no invocation recorded.

multi_head_attention_forward Not yet run
unverified (harvested, no recorded run)
atten4vis/lw-detr/models/attention.py
code served (permissive licence) · get_code("57e6ba78fde593e0")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 2 the same month, 2 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
sigmoid_focal_loss 56 other papers
46 dated earlier, 10 later · 1 only as an assembly brick
2005.12872 (2020-05)
End-to-End Object Detection with Transformers
2010.04159 (2020-10)
Deformable DETR: Deformable Transformers for End-to-End Object Detection
2103.11731 (2021-03)
Meta-DETR: Image-Level Few-Shot Object Detection with Inter-Class Correlation Exploitation
and 53 more not named here
convert_to_xywh 24 other papers
22 dated earlier, 1 the same month, 1 same year, order not determinable
1512.02325 (2015-12)
SSD: Single Shot MultiBox Detector
2010.04159 (2020-10)
Deformable DETR: Deformable Transformers for End-to-End Object Detection
2103.11731 (2021-03)
Meta-DETR: Image-Level Few-Shot Object Detection with Inter-Class Correlation Exploitation
and 21 more not named here
evaluate 23 other papers
21 dated earlier, 1 the same month, 1 same year, order not determinable
1512.02325 (2015-12)
SSD: Single Shot MultiBox Detector
2010.04159 (2020-10)
Deformable DETR: Deformable Transformers for End-to-End Object Detection
2103.11731 (2021-03)
Meta-DETR: Image-Level Few-Shot Object Detection with Inter-Class Correlation Exploitation
and 20 more not named here
hflip 10 other papers
8 dated earlier, 2 later
2106.00666 (2021-06)
You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection
2112.05375 (2021-12)
Rethinking the Two-Stage Framework for Grounded Situation Recognition
2112.08647 (2021-12)
QAHOI: Query-Based Anchors for Human-Object Interaction Detection
and 7 more not named here
crop 5 other papers
5 dated earlier
2106.00666 (2021-06)
You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection
2112.05375 (2021-12)
Rethinking the Two-Stage Framework for Grounded Situation Recognition
2112.08647 (2021-12)
QAHOI: Query-Based Anchors for Human-Object Interaction Detection
and 2 more not named here

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

In this paper, we present a light-weight detection transformer, LW-DETR, which outperforms YOLOs for real-time object detection. The architecture is a simple stack of a ViT encoder, a projector, and a shallow DETR decoder. Our approach leverages recent advanced techniques, such as training-effective techniques, e.g., improved loss and pretraining, and interleaved window and global attentions for reducing the ViT encoder complexity. We improve the ViT encoder by aggregating multi-level feature maps, and the intermediate and final feature maps in the ViT encoder, forming richer feature maps, and introduce window-major feature map organization for improving the efficiency of interleaved attention computation. Experimental results demonstrate that the proposed approach is superior over existing real-time detectors, e.g., YOLO and its variants, on COCO and other benchmark datasets. Code and models are available at (https://github.com/Atten4Vis/LW-DETR).

For agents

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

get_harvested_code_for_paper("2406.03459")
get_code_for_paper("2406.03459")
have("2406.03459")

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

curl https://syntology.ai/api/ran/2406.03459.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.03459.svg)](https://syntology.ai/paper/2406.03459)

Connect an agent — have() is free.