SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.10191 · CVPR · 2024

Generative Region-Language Pretraining for Open-Ended Object Detection

Yi Jiang, Zehuan Yuan, Jianfei Cai, Lizhen Qu, Chuang Lin

arXiv · PDF · Open in the Atlas

Code that ran

8 of 11 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, 0 violated it and 0 have no contract verdict; 8 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/2403.10191.json · badge: /api/ran/2403.10191.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
FoundationVision/GenerateU canonical (classifier label)
classifier basis: census
8 ran · 3 no recorded run
FunctionStatusWhere it lives
batched_nms Ran
ran; our contract draft was wrong, not the code
FoundationVision/GenerateU/detectron2/layers/nms.py
pointer only (licence: MIT); the stored copy get_code serves, from thinklab-sjtu/bench2drivezoo, is pointer only (licence: NOASSERTION) · get_code("d7527a665dfb8f27")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from thinklab-sjtu/bench2drivezoo, 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.

ciou_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
FoundationVision/GenerateU/detectron2/layers/losses.py
code served (permissive licence) · get_code("cee8870d4fbccc39")
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 1, dtype family float, shape (2,) 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 two sets of bounding boxes in XYXY format
# boxes1: shape (N, 4) where N=2
# Each box is [x1, y1, x2, y2] where x2 >= x1 and y2 >= y1
boxes1 = torch.tensor([
    [10.0, 20.0, 50.0, 80.0],  # box 1: x1=10, y1=20, x2=50, y2=80
    [5.0, 10.0, 25.0, 40.0]     # box 2: x1=5, y1=10, x2=25, y2=40
])

boxes2 = torch.tensor([
    [15.0, 25.0, 55.0, 85.0],  # box 1: x1=15, y1=25, x2=55, y2=85
    [8.0, 12.0, 30.0, 45.0]     # box 2: x1=8, y1=12, x2=30, y2=45
])

reduction = 'none'
eps = 1e-07

ARGS = {
    'boxes1': boxes1,
    'boxes2': boxes2,
    'reduction': reduction,
    'eps': eps
}
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 1, dtype family float, shape (2,) — shape (2,) 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
convert_basic_c2_names Ran
ran; our contract draft was wrong, not the code
FoundationVision/GenerateU/detectron2/checkpoint/c2_model_loading.py
code served (permissive licence) · get_code("e8526a516b4f9206")
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 copy
import re

# Analyze the function: it takes a list of strings (original_keys) and applies
# various string transformations to them. It doesn't use any complex objects,
# just list operations and string manipulations.

ARGS = {
    'original_keys': [
        'pred_b',
        'pred_w',
        'conv_b',
        'conv_w',
        'bn_s',
        'bn_bias',
        'bn_rm',
        'bn_riv',
        'bn_gamma',
        'bn_beta',
        'gn_s',
        'gn_bias',
        'res_conv1_norm_weight',
        'conv1_weight',
        'layer_branch1_weight',
        'layer_branch2a_bias',
        'layer_branch2b_weight',
        'layer_branch2c_bias',
        'body_conv_fcn_weight',
        'AnnIndex_lowres',
        'Index_UV_lowres',
        'U_lowres',
        'V_lowres',
    ]
}
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
convert_c2_detectron_names Ran
ran; our contract draft was wrong, not the code
FoundationVision/GenerateU/detectron2/checkpoint/c2_model_loading.py
code served (permissive licence) · get_code("e36f87059208bb7e")
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 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 copy
import logging
import re
import torch

torch.manual_seed(0)

# Create a small dictionary of weights simulating Caffe2 model weights
weights = {
    'conv1_w': torch.randn(3, 3, 3, 3),
    'conv1_b': torch.randn(3),
    'res_conv1_norm_s': torch.randn(3),
    'res_conv1_norm_bias': torch.randn(3),
    'res2_0_branch1_w': torch.randn(4, 3, 1, 1),
    'res2_0_branch1_b': torch.randn(4),
    'res2_0_branch2a_w': torch.randn(4, 3, 3, 3),
    'res2_0_branch2a_b': torch.randn(4),
    'res2_0_branch2b_w': torch.randn(4, 4, 3, 3),
    'res2_0_branch2b_b': torch.randn(4),
    'res2_0_branch2c_w': torch.randn(8, 4, 1, 1),
    'res2_0_branch2c_b': torch.randn(8),
    'bn_s': torch.randn(3),
    'bn_bias': torch.randn(3),
    'bn_rm': torch.randn(3),
    'bn_riv': torch.randn(3),
    'pred_w': torch.randn(10, 8),
    'pred_b': torch.randn(10),
    'fpn_inner_res2_2_sum_lateral_w': torch.randn(4, 8, 1, 1),
    'fpn_inner_res2_2_sum_lateral_b': torch.randn(4),
    'fpn_res2_2_sum_w': torch.randn(4, 4, 3, 3),
    'fpn_res2_2_sum_b': torch.randn(4),
    'rpn_cls_logits_w': torch.randn(3, 4, 3, 3),
    'rpn_cls_logits_b': torch.randn(3),
    'rpn_bbox_pred_w': torch.randn(12, 4, 3, 3),
    'rpn_bbox_pred_b': torch.randn(12),
    'fc6_w': torch.randn(16, 8),
    'fc6_b': torch.randn(16),
    'fc7_w': torch.randn(16, 16),
    'fc7_b': torch.randn(16),
}

ARGS = {
    'weights': weights,
}
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 1, 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
diou_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
FoundationVision/GenerateU/detectron2/layers/losses.py
code served (permissive licence) · get_code("26f3066ead0d2eec")
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 1, dtype family float, shape (2,) 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 sets of bounding boxes in XYXY format
# boxes1: shape (N, 4) where N=2
# Each box is [x1, y1, x2, y2] where x2 >= x1 and y2 >= y1
boxes1 = torch.tensor([
    [10.0, 20.0, 50.0, 60.0],  # box 1: valid box
    [5.0, 10.0, 25.0, 40.0]     # box 2: valid box
])

boxes2 = torch.tensor([
    [15.0, 25.0, 55.0, 65.0],  # box 1: overlaps with boxes1[0]
    [20.0, 15.0, 45.0, 50.0]   # box 2: overlaps with boxes1[1]
])

reduction = 'none'
eps = 1e-07

ARGS = {
    'boxes1': boxes1,
    'boxes2': boxes2,
    'reduction': reduction,
    'eps': eps
}
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 1, dtype family float, shape (2,) — shape (2,) 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
pad_masks Ran
ran; our contract draft was wrong, not the code
FoundationVision/GenerateU/detectron2/layers/mask_ops.py
pointer only (licence: MIT); the stored copy get_code serves, from thinklab-sjtu/bench2drivezoo, is pointer only (licence: NOASSERTION) · get_code("406f84cf799ed34d")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from thinklab-sjtu/bench2drivezoo, 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.

paste_mask_in_image_old Ran
ran; our contract draft was wrong, not the code
FoundationVision/GenerateU/detectron2/layers/mask_ops.py
pointer only (licence: MIT); the stored copy get_code serves, from thinklab-sjtu/bench2drivezoo, is pointer only (licence: NOASSERTION) · get_code("d03592674030077f")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from thinklab-sjtu/bench2drivezoo, 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.

paste_masks_in_image Ran
ran; our contract draft was wrong, not the code
FoundationVision/GenerateU/detectron2/layers/mask_ops.py
pointer only (licence: MIT); the stored copy get_code serves, from thinklab-sjtu/bench2drivezoo, is pointer only (licence: NOASSERTION) · get_code("c0eac7dcd1335fa1")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from thinklab-sjtu/bench2drivezoo, 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.

align_and_update_state_dicts Not yet run
unverified (harvested, no recorded run)
FoundationVision/GenerateU/detectron2/checkpoint/c2_model_loading.py
code served (permissive licence) · get_code("ac2638463572c152")
How it ran

no invocation recorded.

batched_nms_rotated Not yet run
unverified (harvested, no recorded run)
FoundationVision/GenerateU/detectron2/layers/nms.py
code served (permissive licence) · get_code("627f9dec92f2462b")
How it ran

no invocation recorded.

nms_rotated Not yet run
unverified (harvested, no recorded run)
FoundationVision/GenerateU/detectron2/layers/nms.py
code served (permissive licence) · get_code("d15c7cc7a28e4a67")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 11 are also present in other papers whose code we hold (11 with at least one paper dated earlier, 3 the same month, 10 later) and 0 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
convert_basic_c2_names 45 other papers
40 dated earlier, 1 the same month, 4 later
1703.06870 (2017-03)
Mask R-CNN
1804.08864 (2018-04)
Learning to See the Invisible: End-to-End Trainable Amodal Instance Segmentation
1901.08164 (2019-01)
Decoupled Greedy Learning of CNNs
and 42 more not named here
pad_masks 22 other papers
19 dated earlier, 1 the same month, 2 later
1703.06870 (2017-03)
Mask R-CNN
1804.08864 (2018-04)
Learning to See the Invisible: End-to-End Trainable Amodal Instance Segmentation
1908.07836 (2019-08)
PubLayNet: largest dataset ever for document layout analysis
and 19 more not named here
paste_mask_in_image_old 22 other papers
19 dated earlier, 1 the same month, 2 later
1703.06870 (2017-03)
Mask R-CNN
1804.08864 (2018-04)
Learning to See the Invisible: End-to-End Trainable Amodal Instance Segmentation
1908.07836 (2019-08)
PubLayNet: largest dataset ever for document layout analysis
and 19 more not named here
align_and_update_state_dicts 13 other papers
10 dated earlier, 3 later
1912.08193 (2019-12)
PointRend: Image Segmentation as Rendering
2211.09808 (2022-11)
Uni-Perceiver v2: A Generalist Model for Large-Scale Vision and Vision-Language Tasks
and 10 more not named here
convert_c2_detectron_names 13 other papers
10 dated earlier, 3 later
1912.08193 (2019-12)
PointRend: Image Segmentation as Rendering
2211.09808 (2022-11)
Uni-Perceiver v2: A Generalist Model for Large-Scale Vision and Vision-Language Tasks
and 10 more not named here
batched_nms_rotated 4 other papers
3 dated earlier, 1 later
2210.11035 (2022-10)
PointTAD: Multi-Label Temporal Action Detection with Learnable Query Points
2304.03752 (2023-04)
V3Det: Vast Vocabulary Visual Detection Dataset
2307.13619 (2023-07)
RecursiveDet: End-to-End Region-based Recursive Object Detection
and 1 more not named here
ciou_loss 4 other papers
3 dated earlier, 1 later
1708.02002 (2017-08)
Focal Loss for Dense Object Detection
2304.03752 (2023-04)
V3Det: Vast Vocabulary Visual Detection Dataset
2312.08875 (2023-12)
What, How, and When Should Object Detectors Update in Continually Changing Test Domains?
and 1 more not named here
diou_loss 4 other papers
3 dated earlier, 1 later
1708.02002 (2017-08)
Focal Loss for Dense Object Detection
2304.03752 (2023-04)
V3Det: Vast Vocabulary Visual Detection Dataset
2312.08875 (2023-12)
What, How, and When Should Object Detectors Update in Continually Changing Test Domains?
and 1 more not named here
batched_nms 3 other papers
2 dated earlier, 1 later
2304.03752 (2023-04)
V3Det: Vast Vocabulary Visual Detection Dataset
2312.08875 (2023-12)
What, How, and When Should Object Detectors Update in Continually Changing Test Domains?
2406.03877 (2024-06)
Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Driving
paste_masks_in_image 3 other papers
2 dated earlier, 1 later
2304.03752 (2023-04)
V3Det: Vast Vocabulary Visual Detection Dataset
2312.08875 (2023-12)
What, How, and When Should Object Detectors Update in Continually Changing Test Domains?
2406.03877 (2024-06)
Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Driving
nms_rotated 1 other paper
1 dated earlier
2304.03752 (2023-04)
V3Det: Vast Vocabulary Visual Detection Dataset

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 recent research, significant attention has been devoted to the open-vocabulary object detection task, aiming to generalize beyond the limited number of classes labeled during training and detect objects described by arbitrary category names at inference. Compared with conventional object detection, open vocabulary object detection largely extends the object detection categories. However, it relies on calculating the similarity between image regions and a set of arbitrary category names with a pretrained visionand-language model. This implies that, despite its open-set nature, the task still needs the predefined object categories during the inference stage. This raises the question: What if we do not have exact knowledge of object categories during inference? In this paper, we call such a new setting as generative open-ended object detection, which is a more general and practical problem. To address it, we formulate object detection as a generative problem and propose a simple framework named GenerateU, which can detect dense objects and generate their names in a free-form way. Particularly, we employ Deformable DETR as a region proposal generator with a language model translating visual regions to object names. To assess the free-form object detection task, we introduce an evaluation method designed to quantitatively measure the performance of generative outcomes. Extensive experiments demonstrate strong zero-shot detection performance of our GenerateU. For example, on the LVIS dataset, our GenerateU achieves comparable results to the open-vocabulary object detection method GLIP, even though the category names are not seen by Genera-teU during inference. Code is available at: https:// github.com/FoundationVision/GenerateU .

For agents

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

get_harvested_code_for_paper("2403.10191")
get_code_for_paper("2403.10191")
have("2403.10191")

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

curl https://syntology.ai/api/ran/2403.10191.json

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

[![Syntology run record](https://syntology.ai/api/ran/2403.10191.svg)](https://syntology.ai/paper/2403.10191)

Connect an agent — have() is free.