SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · aaai_28994 · AAAI · 2024

Mixup-Induced Domain Extrapolation for Domain Generalization

Songcan Chen, Meng Cao

Open in the Atlas

Code that ran

8 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, 0 violated it and 8 have no contract verdict; 0 failures of our instrument; 4 no recorded run), measured 2026-11-22; 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/aaai_28994.json · badge: /api/ran/aaai_28994.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
Alrash/EDM canonical (classifier label)
classifier basis: census
8 ran · 4 no recorded run
FunctionStatusWhere it lives
Nmax Ran
ran on a synthesized input
Alrash/EDM/src/DeepDG/datautil/util.py
code served (permissive licence) · get_code("b6e05a6d369ffe88")
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 int, ndim 0, dtype family int, 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 numpy as np

torch.manual_seed(0)

# The function Nmax iterates through test_envs checking if d < test_envs[i]
# test_envs should be a list/sequence of comparable numeric values
# d should be a numeric value comparable to elements in test_envs

test_envs = [10, 20, 30, 40, 50]
d = 25

ARGS = {
    "test_envs": test_envs,
    "d": d
}
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 int, ndim 0, dtype family int, 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
accuracy Ran
ran on a synthesized input
Alrash/EDM/src/DeepDG/alg/modelopera.py
code served (permissive licence) · get_code("e31d1f2fae01ad6e")
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 float, ndim 0, dtype family float, shape () 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
import torch.nn as nn

torch.manual_seed(0)

class SimpleNetwork(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def predict(self, x):
        return self.linear(x)

network = SimpleNetwork()

loader = [
    (torch.randn(2, 4), torch.randint(0, 3, (2,))),
    (torch.randn(2, 4), torch.randint(0, 3, (2,))),
]

ARGS = {
    "network": network,
    "loader": loader,
}
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 float, 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
calc_normal_distribution Ran
ran on a synthesized input
Alrash/EDM/mix.py
code served (permissive licence) · get_code("608eca2e5decc608")
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 torch

torch.manual_seed(0)

# x is a tensor that gets indexed by d and has mean/std computed along certain axes
# For 4D tensor: axes are [0, 2, 3], for other: axes are [0]
# x needs to support indexing with a boolean mask from d
x = torch.randn(2, 4, 3, 5)  # 4D tensor: (batch=2, channels=4, height=3, width=5)

# d is a tensor of domain labels that:
# - gets compared with == to create boolean masks
# - has .unique() method called on it
# - is used to index x
# d should have same first dimension as x (batch dimension)
d = torch.tensor([0, 1])  # domain labels for each sample in batch

ARGS = {
    "x": x,
    "d": d,
}
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
get_optimizer Ran
ran on a synthesized input
Alrash/EDM/src/DeepDG/alg/opt.py
code served (permissive licence) · get_code("9c0835032674dc95")
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 SGD. 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)

# Create minimal nn.Module classes with required attributes/methods
class SimpleFeaturizer(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.fc(x)

class SimpleClassifier(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc = nn.Linear(3, 2)
    
    def forward(self, x):
        return self.fc(x)

class SimpleDiscriminator(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc = nn.Linear(3, 1)
    
    def forward(self, x):
        return self.fc(x)

class SimpleClassEmbeddings(nn.Module):
    def __init__(self):
        super().__init__()
        self.emb = nn.Embedding(2, 3)
    
    def forward(self, x):
        return self.emb(x)

class AlgorithmModel(nn.Module):
    def __init__(self, include_discriminator=False, include_class_embeddings=False):
        super().__init__()
        self.featurizer = SimpleFeaturizer()
        self.classifier = SimpleClassifier()
        if include_discriminator:
            self.discriminator = SimpleDiscriminator()
        if include_class_embeddings:
            self.class_embeddings = SimpleClassEmbeddings()

class Args:
    def __init__(self):
        self.lr = 0.01
        self.inner_lr = 0.005
        self.lr_decay1 = 0.1
        self.lr_decay2 = 0.01
        self.momentum = 0.9
        self.weight_decay = 1e-4
        self.schuse = False
        self.schusech = 'cos'
        self.algorithm = 'ERM'

# Create test algorithm model
alg = AlgorithmModel(include_discriminator=False, include_class_embeddings=False)

# Create args object
args = Args()

ARGS = {
    'alg': alg,
    'args': args,
    'inner': False,
    'alias': True,
    'isteacher': 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 SGD — a SGD output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SGD 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
get_params Ran
ran on a synthesized input
Alrash/EDM/src/DeepDG/alg/opt.py
code served (permissive licence) · get_code("ecd2696484224277")
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 torch
import torch.nn as nn

torch.manual_seed(0)

class SimpleFeaturizer(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc = nn.Linear(4, 8)
    
    def forward(self, x):
        return self.fc(x)

class SimpleClassifier(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc = nn.Linear(8, 2)
    
    def forward(self, x):
        return self.fc(x)

class SimpleDiscriminator(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc = nn.Linear(8, 1)
    
    def forward(self, x):
        return self.fc(x)

class SimpleClassEmbeddings(nn.Module):
    def __init__(self):
        super().__init__()
        self.embeddings = nn.Embedding(2, 8)
    
    def forward(self, x):
        return self.embeddings(x)

class SimpleAlgorithm(nn.Module):
    def __init__(self, algorithm_name):
        super().__init__()
        self.featurizer = SimpleFeaturizer()
        self.classifier = SimpleClassifier()
        self.discriminator = SimpleDiscriminator()
        self.class_embeddings = SimpleClassEmbeddings()
        self.algorithm_name = algorithm_name
    
    def forward(self, x):
        feat = self.featurizer(x)
        out = self.classifier(feat)
        return out

class SimpleAlgorithmNoAlias(nn.Module):
    def __init__(self, algorithm_name):
        super().__init__()
        self.module_list = nn.ModuleList([SimpleFeaturizer(), SimpleClassifier()])
        self.discriminator = SimpleDiscriminator()
        self.class_embeddings = SimpleClassEmbeddings()
        self.algorithm_name = algorithm_name
    
    def __getitem__(self, idx):
        return self.module_list[idx]

class SimpleAlgorithmTeacher(nn.Module):
    def __init__(self):
        super().__init__()
        self.module_list = nn.ModuleList([
            SimpleFeaturizer(),
            SimpleClassifier(),
            SimpleClassifier()
        ])
    
    def __getitem__(self, idx):
        return self.module_list[idx]

class SimpleArgs:
    def __init__(self, algorithm='ERM', schuse=False, schusech='cos', lr=0.01, 
                 inner_lr=0.001, lr_decay1=1.0, lr_decay2=0.5):
        self.algorithm = algorithm
        self.schuse = schuse
        self.schusech = schusech
        self.lr = lr
        self.inner_lr = inner_lr
        self.lr_decay1 = lr_decay1
        self.lr_decay2 = lr_decay2

# Test case 1: Basic case with alias=True
alg1 = SimpleAlgorithm('ERM')
args1 = SimpleArgs(algorithm='ERM', schuse=False, lr=0.01, lr_decay1=1.0, lr_decay2=0.5)

# Test case 2: With DANN algorithm
alg2 = SimpleAlgorithm('DANN')
args2 = SimpleArgs(algorithm='DANN', schuse=False, lr=0.01, lr_decay1=1.0, lr_decay2=0.5)

# Test case 3: With CDANN algorithm
alg3 = SimpleAlgorithm('CDANN')
args3 = SimpleArgs(algorithm='CDANN', schuse=False, lr=0.01, lr_decay1=1.0, lr_decay2=0.5)

# Test case 4: Inner loop
alg4 = nn.ModuleList([SimpleFeaturizer(), SimpleClassifier()])
args4 = SimpleArgs(algorithm='ERM', schuse=False, inner_lr=0.001, lr_decay1=1.0, lr_decay2=0.5)

# Test case 5: Teacher
alg5 = SimpleAlgorithmTeacher()
args5 = SimpleArgs(algorithm='ERM', schuse=False, lr=0.01, lr_decay1=1.0, lr_decay2=0.5)

# Test case 6: With scheduler
alg6 = SimpleAlgorithm('ERM')
args6 = SimpleArgs(algorithm='ERM', schuse=True, schusech='cos', lr=0.01, lr_decay1=1.0, lr_decay2=0.5)

ARGS = {
    'alg': alg1,
    'args': args1,
    'inner': False,
    'alias': True,
    'isteacher': 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 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
projection Ran
ran on a synthesized input
Alrash/EDM/mix.py
code served (permissive licence) · get_code("2b8ab603a6727f76")
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)

# The projection function performs normalization and denormalization
# It takes 5 parameters: x, mu, std, mixed_mu, mixed_std
# All are used in arithmetic operations (subtraction, division, multiplication, addition)
# They should all be tensors of compatible shapes for broadcasting

x = torch.randn(2, 4)
mu = torch.randn(2, 4)
std = torch.randn(2, 4)
mixed_mu = torch.randn(2, 4)
mixed_std = torch.randn(2, 4)

ARGS = {
    "x": x,
    "mu": mu,
    "std": std,
    "mixed_mu": mixed_mu,
    "mixed_std": mixed_std,
}
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
random_pairs_of_minibatches Ran
ran on a synthesized input
Alrash/EDM/src/DeepDG/datautil/util.py
code served (permissive licence) · get_code("fff95ee8729333e4")
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. 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)

# Create a minimal args object with batch_size attribute
class Args:
    def __init__(self):
        self.batch_size = 2

args = Args()

# Create minibatches as a list of [x, y, d] tuples
# Each minibatch contains: images (batch_size, channels, height, width), labels, domain labels
minibatch1 = [
    torch.randn(2, 3, 4, 4),  # x: batch_size=2, 3 channels, 4x4 spatial
    torch.tensor([0, 1]),      # y: labels
    torch.tensor([0, 0])       # d: domain labels
]

minibatch2 = [
    torch.randn(2, 3, 4, 4),  # x: batch_size=2, 3 channels, 4x4 spatial
    torch.tensor([1, 0]),      # y: labels
    torch.tensor([1, 1])       # d: domain labels
]

minibatches = [minibatch1, minibatch2]

# Create augmented data (optional)
# aug = [aug_x, aug_y, aug_d] where each has total 4 samples (2 minibatches * batch_size 2)
aug = [
    torch.randn(4, 3, 4, 4),   # augmented x: 4 samples
    torch.tensor([0, 1, 1, 0]), # augmented y: 4 labels
    torch.tensor([0, 0, 1, 1])  # augmented d: 4 domain labels
]

ARGS = {
    "args": args,
    "minibatches": minibatches,
    "aug": aug
}
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 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
random_pairs_of_minibatches_by_domainperm Ran
ran on a synthesized input
Alrash/EDM/src/DeepDG/datautil/util.py
code served (permissive licence) · get_code("2fac89dc34dc19ce")
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. 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)

# The function expects minibatches: a list of tuples where each tuple is (x, y)
# x and y are tensors or array-like objects that support len() and slicing

minibatches = [
    (torch.randn(5, 4), torch.randint(0, 2, (5,))),  # minibatch 1: 5 samples, 4 features
    (torch.randn(6, 4), torch.randint(0, 2, (6,))),  # minibatch 2: 6 samples, 4 features
    (torch.randn(4, 4), torch.randint(0, 2, (4,))),  # minibatch 3: 4 samples, 4 features
]

ARGS = {
    "minibatches": minibatches
}
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
calc_coeff Not yet run
unverified (harvested, no recorded run)
Alrash/EDM/src/DeepDG/network/util.py
pointer only (licence: MIT); the stored copy get_code serves, from Yiyang98/BFUDA, is pointer only (licence: NONE) · get_code("e352afa5762c5a6b")
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 Yiyang98/BFUDA, 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. No invocation recorded.

calc_momentum_distribution_parameters Not yet run
unverified (harvested, no recorded run)
Alrash/EDM/mix.py
code served (permissive licence) · get_code("dfa5594697bc8ad1")
How it ran

no invocation recorded.

get_algorithm_class Not yet run
unverified (harvested, no recorded run)
Alrash/EDM/src/DeepDG/alg/alg.py
code served (permissive licence) · get_code("2676a45c41ebada6")
How it ran

no invocation recorded.

get_scheduler Not yet run
unverified (harvested, no recorded run)
Alrash/EDM/src/DeepDG/alg/opt.py
code served (permissive licence) · get_code("8e47b3c4f99ce70e")
How it ran

no invocation recorded.

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, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 1 later) and 9 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
get_algorithm_class 39 other papers
34 dated earlier, 1 later, 4 same year, order not determinable
1907.02893 (2019-07)
Invariant Risk Minimization
2007.01434 (2020-07)
In Search of Lost Domain Generalization
2007.02931 (2020-07)
Adaptive Risk Minimization: Learning to Adapt to Domain Shift
and 36 more not named here
calc_coeff 14 other papers
11 dated earlier, 3 same year, order not determinable · 2 only as an assembly brick
1505.07818 (2015-05)
Domain-Adversarial Training of Neural Networks
1904.06487 (2019-04)
Semi-supervised Domain Adaptation via Minimax Entropy
2003.02541 (2020-03)
A Balanced and Uncertainty-aware Approach for Partial Domain Adaptation
and 11 more not named here
accuracy 1 other paper
1 dated earlier
2201.06648 (2022-01)
OmniPrint: A Configurable Printed Character Synthesizer

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

Domain generalization aims to learn a well-performed classifier on multiple source domains for unseen target domains under domain shift. Domain-invariant representation (DIR) is an intuitive approach and has been of great concern. In practice, since the targets are variant and agnostic, only a few sources are not sufficient to reflect the entire domain population, leading to biased DIR. Derived from PAC-Bayes framework, we provide a novel generalization bound involving the number of domains sampled from the environment (N ) and the radius of the Wasserstein ball centred on the target (r), which have rarely been considered before. Herein, we can obtain two natural and significant findings: when N increases, 1) the gap between the source and target sampling environments can be gradually mitigated; 2) the target can be better approximated within the Wasserstein ball. These findings prompt us to collect adequate domains against domain shift. For seeking convenience, we design a novel yet simple Extrapolation Domain strategy induced by the Mixup scheme, namely EDM. Through a reverse Mixup scheme to generate the extrapolated domains, combined with the interpolated domains, we expand the interpolation space spanned by the sources, providing more abundant domains to increase sampling intersections to shorten r. Moreover, EDM is easy to implement and be plugged-and-played. In experiments, EDM has been plugged into several methods in both closed and open set settings, achieving up to 5.73% improvement.

For agents

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

get_harvested_code_for_paper("aaai_28994")
get_code_for_paper("aaai_28994")
have("aaai_28994")

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

curl https://syntology.ai/api/ran/aaai_28994.json

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

[![Syntology run record](https://syntology.ai/api/ran/aaai_28994.svg)](https://syntology.ai/paper/aaai_28994)

Connect an agent — have() is free.