SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.10918 · 2024

PartImageNet++ Dataset: Scaling up Part-based Models for Robust Recognition

arXiv · PDF · Open in the Atlas

Code that ran

10 of 19 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; 10 failures of our instrument; 9 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/2407.10918.json · badge: /api/ran/2407.10918.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
LixiaoTHU/PartImageNetPP canonical (classifier label)
classifier basis: census
10 ran · 9 no recorded run
FunctionStatusWhere it lives
build_loss Ran
ran; our contract draft was wrong, not the code
LixiaoTHU/PartImageNetPP/model/loss.py
code served (permissive licence) · get_code("c8696b362bb9873b")
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. 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 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
from timm.loss import JsdCrossEntropy, BinaryCrossEntropy, LabelSmoothingCrossEntropy, SoftTargetCrossEntropy

torch.manual_seed(0)

# Stand-in class for args
class Args:
    def __init__(self, smoothing=0.0):
        self.smoothing = smoothing

# Test case 1: with mixup_fn (should use SoftTargetCrossEntropy)
class SimpleMixupFn:
    def __call__(self, x, y):
        return x, y

# Create different argument configurations
ARGS = {
    "args": Args(smoothing=0.1),
    "mixup_fn": SimpleMixupFn()
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
timm, 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
build_seg_loss Ran
ran; our contract draft was wrong, not the code
LixiaoTHU/PartImageNetPP/model/loss.py
code served (permissive licence) · get_code("372193f48cfccb0d")
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. 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 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)

class MockMixupFn:
    def __call__(self, x, y):
        return x, y

class MockArgs:
    def __init__(self):
        self.dataset_version = 'single'

ARGS = {
    'args': MockArgs(),
    'mixup_fn': MockMixupFn()
}
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
build_transform Ran
ran; our contract draft was wrong, not the code
LixiaoTHU/PartImageNetPP/eval_multi_dataset.py
code served (permissive licence) · get_code("123de186a05d2490")
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 Compose. 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 torchvision import transforms
from torchvision.transforms import InterpolationMode

class Args:
    def __init__(self, input_size, crop_pct):
        self.input_size = input_size
        self.crop_pct = crop_pct

ARGS = {
    "args": Args(input_size=224, crop_pct=0.875)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Compose — a Compose output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Compose 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
conv1x1 Ran
ran; our contract draft was wrong, not the code
LixiaoTHU/PartImageNetPP/model/resnet.py
code served (permissive licence) · get_code("d9def42110729a85")
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
import torch.nn as nn

torch.manual_seed(0)

# The function conv1x1 takes in_planes, out_planes, and stride as parameters
# It returns an nn.Conv2d layer configured for 1x1 convolution
# We need to provide valid integer arguments for these parameters

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
conv3x3 Ran
ran; our contract draft was wrong, not the code
LixiaoTHU/PartImageNetPP/model/resnet.py
code served (permissive licence) · get_code("160bb14bd76201b4")
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
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "in_planes": 3,
    "out_planes": 64,
    "stride": 1,
    "groups": 1,
    "dilation": 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
downsample_seglabel Ran
ran; our contract draft was wrong, not the code
LixiaoTHU/PartImageNetPP/model/utils.py
code served (permissive licence) · get_code("7c2b72fae4adf239")
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
import torch.nn.functional as F

torch.manual_seed(0)

class ArgsConfig:
    def __init__(self):
        self.seg_classes = 5
        self.output_size = [8, 4]

ori_seg = torch.randint(0, 5, (2, 1, 16, 16)).float()

args = ArgsConfig()

ARGS = {
    'ori_seg': ori_seg,
    'args': args,
    'return_pixel_seg': 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 — 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
replace_best Ran
ran; our contract draft was wrong, not the code
LixiaoTHU/PartImageNetPP/adv_utils.py
code served (permissive licence) · get_code("d5138507f6a88870")
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)

# Construct arguments for replace_best function
# The function expects:
# - loss: tensor of losses
# - bloss: tensor of best losses (or None)
# - x: tensor of current values
# - bx: tensor of best values (or None)

# Create small tensors with batch size 2
loss = torch.tensor([0.5, 0.3], dtype=torch.float32)
bloss = torch.tensor([0.6, 0.2], dtype=torch.float32)
x = torch.randn(2, 4, dtype=torch.float32)
bx = torch.randn(2, 4, dtype=torch.float32)

ARGS = {
    'loss': loss,
    'bloss': bloss,
    'x': x,
    'bx': bx
}
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
replace_best_reverse Ran
ran; our contract draft was wrong, not the code
LixiaoTHU/PartImageNetPP/adv_utils.py
code served (permissive licence) · get_code("68ba7f3cb55ef939")
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)

# Construct valid arguments for replace_best_reverse
# The function expects:
# - loss: a tensor of losses (current batch losses)
# - bloss: a tensor of best losses (or None)
# - x: a tensor of current values
# - bx: a tensor of best values

# Create small tensors with batch size 2 and dimension 4
loss = torch.randn(2, 4)
bloss = torch.randn(2, 4)
x = torch.randn(2, 4)
bx = torch.randn(2, 4)

ARGS = {
    "loss": loss,
    "bloss": bloss,
    "x": x,
    "bx": bx
}
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
unwrap_checkpoint Ran
ran; our contract draft was wrong, not the code
LixiaoTHU/PartImageNetPP/utils.py
code served (permissive licence) · get_code("58875d9d3a043ec1")
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 torch.nn as nn

torch.manual_seed(0)

# The function unwrap_checkpoint takes a checkpoint dict and extracts the model state
# It checks for 'state_dict_ema', then 'model', then returns the checkpoint as-is
# So we need to provide a checkpoint dict that could have these keys

# Create a simple checkpoint dict with 'state_dict_ema' key
checkpoint_with_ema = {
    'state_dict_ema': {
        'layer.weight': torch.randn(4, 3),
        'layer.bias': torch.randn(4)
    },
    'other_key': 'some_value'
}

ARGS = {
    'checkpoint': checkpoint_with_ema
}
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 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
validate Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
LixiaoTHU/PartImageNetPP/eval_multi_dataset.py
code served (permissive licence) · get_code("3222066805488570")
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 OrderedDict. 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
import torch.nn as nn
from collections import OrderedDict

torch.manual_seed(0)

# Create a minimal model that accepts input and returns logits
class MinimalModel(nn.Module):
    def __init__(self, num_classes=10):
        super().__init__()
        self.fc = nn.Linear(4, num_classes)
    
    def forward(self, x):
        return self.fc(x)

# Create a minimal data loader (list of (input, target) tuples)
def create_loader(batch_size=2, num_batches=3, num_classes=10):
    loader = []
    for _ in range(num_batches):
        # input: (batch_size, 4), target: (batch_size,)
        input_tensor = torch.randn(batch_size, 4)
        target_tensor = torch.randint(0, num_classes, (batch_size,))
        loader.append((input_tensor, target_tensor))
    return loader

# Create args object with required attributes
class Args:
    def __init__(self, num_classes=10):
        self.num_classes = num_classes

# Construct the arguments
model = MinimalModel(num_classes=10)
loader = create_loader(batch_size=2, num_batches=3, num_classes=10)
args = Args(num_classes=10)
classmask = None
log_suffix = 'clean acc'

ARGS = {
    'model': model,
    'loader': loader,
    'args': args,
    'classmask': classmask,
    'log_suffix': log_suffix
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
timm, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type OrderedDict — a OrderedDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a OrderedDict 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
build_model Not yet run
unverified (harvested, no recorded run)
LixiaoTHU/PartImageNetPP/model/model.py
code served (permissive licence) · get_code("9381ca1fec4b4092")
How it ran

no invocation recorded.

build_part_model Not yet run
unverified (harvested, no recorded run)
LixiaoTHU/PartImageNetPP/model/model.py
code served (permissive licence) · get_code("940deb1626cd6b57")
How it ran

no invocation recorded.

clamp Not yet run
unverified (harvested, no recorded run)
LixiaoTHU/PartImageNetPP/adv_utils.py
code served (permissive licence) · get_code("8a93e041134b597a")
How it ran

no invocation recorded.

compute_seg_acc Not yet run
unverified (harvested, no recorded run)
LixiaoTHU/PartImageNetPP/utils.py
code served (permissive licence) · get_code("71dc09fa04061081")
How it ran

no invocation recorded.

create_logger Not yet run
unverified (harvested, no recorded run)
LixiaoTHU/PartImageNetPP/utils.py
code served (permissive licence) · get_code("9955cd681dd897fa")
How it ran

no invocation recorded.

get_filenames_of_category Not yet run
unverified (harvested, no recorded run)
LixiaoTHU/PartImageNetPP/IN_C_helper/wordnet_functions.py
code served (permissive licence) · get_code("64c4b10134d0f5f6")
How it ran

no invocation recorded.

get_hypernyms Not yet run
unverified (harvested, no recorded run)
LixiaoTHU/PartImageNetPP/IN_C_helper/wordnet_functions.py
code served (permissive licence) · get_code("00100b85ac7c0b0f")
How it ran

no invocation recorded.

hypernyms_in_ilsvrc2012_categories Not yet run
unverified (harvested, no recorded run)
LixiaoTHU/PartImageNetPP/IN_C_helper/wordnet_functions.py
code served (permissive licence) · get_code("515a490d85611e63")
How it ran

no invocation recorded.

resnet18 Not yet run
unverified (harvested, no recorded run)
LixiaoTHU/PartImageNetPP/model/resnet.py
code served (permissive licence) · get_code("96581f563c08c5bc")
How it ran

no invocation recorded.

Also present in other papers' code

Of 19 samples lifted from this paper's linked repositories, 19 carry a function identity and 0 do not. Of the 19 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 2 the same month, 4 later) and 15 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
conv1x1 431 other papers
369 dated earlier, 5 the same month, 46 later, 10 same year, order not determinable, 1 undated · 38 only as an assembly brick
1409.1556 (2014-09)
Very Deep Convolutional Networks for Large-Scale Image Recognition
1506.01186 (2015-06)
Cyclical Learning Rates for Training Neural Networks
1507.05717 (2015-07)
An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition
and 428 more not named here
conv3x3 312 other papers
261 dated earlier, 4 the same month, 36 later, 10 same year, order not determinable, 1 undated · 31 only as an assembly brick
1409.1556 (2014-09)
Very Deep Convolutional Networks for Large-Scale Image Recognition
1507.05717 (2015-07)
An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
and 309 more not named here
resnet18 95 other papers
75 dated earlier, 16 later, 3 same year, order not determinable, 1 undated
1805.00907 (2018-05)
Glow: Graph Lowering Compiler Techniques for Neural Networks
1902.08153 (2019-02)
Learned Step Size Quantization
1904.01355 (2019-04)
FCOS: Fully Convolutional One-Stage Object Detection
and 92 more not named here
clamp 33 other papers
30 dated earlier, 3 later · 1 only as an assembly brick
1706.06083 (2017-06)
Towards Deep Learning Models Resistant to Adversarial Attacks
2001.03994 (2020-01)
Fast is better than free: Revisiting adversarial training
2002.11569 (2020-02)
Overfitting in adversarially robust deep learning
and 30 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

Deep learning-based object recognition systems can be easily fooled by various adversarial perturbations. One reason for the weak robustness may be that they do not have part-based inductive bias like the human recognition process. Motivated by this, several part-based recognition models have been proposed to improve the adversarial robustness of recognition. However, due to the lack of part annotations, the effectiveness of these methods is only validated on small-scale nonstandard datasets. In this work, we propose PIN++, short for PartImageNet++, a dataset providing high-quality part segmentation annotations for all categories of ImageNet-1K (IN-1K). With these annotations, we build part-based methods directly on the standard IN-1K dataset for robust recognition. Different from previous two-stage part-based models, we propose a Multi-scale Part-supervised Model (MPM), to learn a robust representation with part annotations. Experiments show that MPM yielded better adversarial robustness on the large-scale IN-1K over strong baselines across various attack settings. Furthermore, MPM achieved improved robustness on common corruptions and several out-of-distribution datasets. The dataset, together with these results, enables and encourages researchers to explore the potential of part-based models in more real applications.

For agents

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

get_harvested_code_for_paper("2407.10918")
get_code_for_paper("2407.10918")
have("2407.10918")

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

curl https://syntology.ai/api/ran/2407.10918.json

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

[![Syntology run record](https://syntology.ai/api/ran/2407.10918.svg)](https://syntology.ai/paper/2407.10918)

Connect an agent — have() is free.