SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.03375 · 2021

Commutative Lie Group VAE for Disentanglement Learning

arXiv · PDF · Open in the Atlas

Code that ran

15 of 15 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 13 have no contract verdict; 2 failures of our instrument; 0 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/2106.03375.json · badge: /api/ran/2106.03375.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
zhuxinqimac/CommutativeLieGroupVAE-Pytorch canonical (classifier label)
classifier basis: census
15 ran · 0 no recorded run
FunctionStatusWhere it lives
attn_stats Ran
ran on a synthesized input
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/models/utils.py
code served (permissive licence) · get_code("c8c0fc37731d8f65")
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
from torch.distributions import Categorical
import collections

torch.manual_seed(0)

# Create small tensors for testing
# attention: shape (batch_size, num_actions) - probability distribution over actions
batch_size = 2
num_actions = 4

attention = torch.softmax(torch.randn(batch_size, num_actions), dim=-1)

# true_actions: can be either 1D (label indices) or 2D (one-hot or multi-dim labels)
# The function sprites_label_to_action handles both cases
# Let's create a 2D case to test the full logic
true_actions = torch.randn(batch_size, num_actions)

ARGS = {
    'attention': attention,
    'true_actions': true_actions
}
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
beta_celeb_encoder Ran
ran on a synthesized input
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/models/beta.py
code served (permissive licence) · get_code("a3fd6939c5efba47")
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 Sequential. 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)

class Args:
    def __init__(self, nc, latents):
        self.nc = nc
        self.latents = latents

ARGS = {
    'args': Args(nc=3, latents=16)
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
beta_shape_encoder Ran
ran on a synthesized input
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/models/beta.py
code served (permissive licence) · get_code("8c9290faed00844f")
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 Sequential. 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)

class Args:
    def __init__(self):
        self.nc = 2
        self.latents = 3

ARGS = {
    "args": Args()
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
beta_shapes_decoder Ran
ran on a synthesized input
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/models/beta.py
code served (permissive licence) · get_code("5a25b9447af5511e")
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 Sequential. 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 a small stand-in config object with the required attributes
class ArgsConfig:
    def __init__(self):
        self.latents = 4  # Small latent dimension
        self.nc = 3  # Number of channels (RGB)

ARGS = {
    'args': ArgsConfig()
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
count_parameters Ran
ran on a synthesized input
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/models/utils.py
code served (permissive licence) · get_code("d2653edfda6250cd")
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 str. 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)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
        self.bn = nn.BatchNorm1d(2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        x = self.bn(x)
        return x

model = SimpleModel()

ARGS = {
    "model": model
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
dip_vae_i_loss Ran
ran on a synthesized input
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/models/dip_vae.py
code served (permissive licence) · get_code("e3db85df9560a4d6")
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

torch.manual_seed(0)

# mu: batch of mean vectors from a VAE
# Shape: (batch_size, latent_dim)
# The function computes covariance statistics, so we need a proper batch
mu = torch.randn(2, 4)

# lambda_d: weight for diagonal regularization (scalar)
lambda_d = 0.5

# lambda_od: weight for off-diagonal regularization (scalar)
lambda_od = 1.0

ARGS = {
    "mu": mu,
    "lambda_d": lambda_d,
    "lambda_od": lambda_od,
}
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
dip_vae_ii_loss Ran
ran on a synthesized input
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/models/dip_vae.py
code served (permissive licence) · get_code("2a32245f4820873d")
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

torch.manual_seed(0)

# Create small tensors for the parameters
batch_size = 2
latent_dim = 4

# mu: mean of the latent distribution, shape (batch_size, latent_dim)
mu = torch.randn(batch_size, latent_dim)

# lv: log variance of the latent distribution, shape (batch_size, latent_dim)
lv = torch.randn(batch_size, latent_dim)

# lambda_d: weight for diagonal regularization (scalar)
lambda_d = 1.0

# lambda_od: weight for off-diagonal regularization (scalar)
lambda_od = 0.5

ARGS = {
    "mu": mu,
    "lv": lv,
    "lambda_d": lambda_d,
    "lambda_od": lambda_od,
}
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
gaussian_kls Ran
ran; our contract draft was wrong, not the code
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/models/losses.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from gomb0c/soft_tpr, is pointer only (licence: NONE) · get_code("406c6e802e6ba500")
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 gomb0c/soft_tpr, 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.

get_mean Ran
ran on a synthesized input
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/collect_results.py
code served (permissive licence) · get_code("a16604e7e8d463ba")
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 float64, ndim 0, 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 numpy as np

# The function get_mean takes a single parameter 'x'
# Looking at the body:
# - x is treated as an iterable (passed to filter and list)
# - filter(None, x) removes falsy values
# - len() is called on the filtered list
# - np.mean() is called on the filtered list
# So x should be a list/array that may contain None or falsy values

ARGS = {
    "x": [1.0, 2.0, None, 3.0, 0, 4.5]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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_num Ran
ran on a synthesized input
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/collect_results.py
code served (permissive licence) · get_code("049a2844665499ee")
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 get_num takes a parameter x and:
# 1. Converts it to a list
# 2. Filters out falsy values (None, 0, False, empty strings, etc.)
# 3. Returns the length of the filtered list

# Valid argument: a list with some truthy and falsy values
x = [1, 2, None, 0, "hello", False, 3.14, ""]

ARGS = {
    "x": x
}
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
get_std Ran
ran on a synthesized input
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/collect_results.py
code served (permissive licence) · get_code("da165531811b0868")
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 float64, ndim 0, 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 numpy as np

def get_std(x):
    x = list(filter(None, x))
    return None if len(x) == 0 else np.std(x)

# The function takes a parameter 'x' which should be iterable
# It filters out None/falsy values and computes standard deviation
# We need to provide a list-like object with some numeric values and possibly some None values

ARGS = {
    'x': [1.0, 2.0, 3.0, 4.0, None, 5.0]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
matrix_diag Ran
ran; our contract draft was wrong, not the code
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/models/dip_vae.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ZhuYun97/RoSA, is pointer only (licence: NONE) · get_code("f29345568db89515")
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 ZhuYun97/RoSA, 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.

parse_val_logs Ran
ran on a synthesized input
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/trainin_loop.py
code served (permissive licence) · get_code("9dd58bd04c5e1e9a")
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)

# Stand-in classes and objects
class Args:
    def __init__(self):
        self.visualise = False

class Logger:
    pass

class State:
    pass

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def imaging_cbs(self, args, logger, model, batch):
        # Returns an iterable of callables
        return []

class VisualizationCallback:
    def __call__(self, model, state, global_step):
        pass

# Build arguments
args = Args()
args.visualise = False

model = SimpleModel()

batch = {
    'x': torch.randn(2, 4),
    'y': torch.randn(2, 3)
}

logger = Logger()

def metric_list_fn():
    return {'loss': 0.5, 'accuracy': 0.95}

metric_list = metric_list_fn

logs = {}

state = State()

t = 0

global_step = 0

ARGS = {
    't': t,
    'args': args,
    'model': model,
    'batch': batch,
    'logger': logger,
    'metric_list': metric_list,
    'logs': logs,
    'state': state,
    'global_step': global_step
}
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
sprites_label_to_action Ran
ran on a synthesized input
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/models/utils.py
code served (permissive licence) · get_code("0b7e7c9fb991cbe8")
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 1, dtype family int, 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)

# Test case 1: 2D label tensor with multiple columns
# Shape: (batch_size=2, num_actions=4)
# This will trigger the main branch: label.shape[-1] > 1 and len(list(label.shape)) > 1
label_2d = torch.tensor([
    [0.5, -0.3, 0.2, 0.1],
    [-0.4, 0.6, -0.5, 0.2]
], dtype=torch.float32)

ARGS = {
    "label": label_2d
}
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 1, dtype family int, 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
to_cuda Ran
ran on a synthesized input
zhuxinqimac/CommutativeLieGroupVAE-Pytorch/trainin_loop.py
code served (permissive licence) · get_code("cfca2b62d69fcc71")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# The function to_cuda() takes one parameter 'x' which can be:
# 1. A tensor (checked via torch.is_tensor)
# 2. An iterable/sequence of nested structures

# Create a test case with nested structures containing tensors
tensor1 = torch.randn(2, 4)
tensor2 = torch.randn(3, 5)
tensor3 = torch.randn(2, 6)

# Test with a nested list structure
nested_structure = [
    tensor1,
    [tensor2, tensor3],
    tensor1
]

ARGS = {
    "x": nested_structure
}
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

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 2 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 2 later) and 13 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
matrix_diag 2 other papers
2 later
2204.13846 (2022-04)
RoSA: A Robust Self-Aligned Framework for Node-Node Graph Contrastive Learning
2301.10677 (2023-01)
Imitating Human Behaviour with Diffusion Models
gaussian_kls 1 other paper
1 later
2412.04671 (2024-12)
Fully Distributed, Flexible Compositional Visual Representations via Soft Tensor Products

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

We view disentanglement learning as discovering an underlying structure that equivariantly reflects the factorized variations shown in data. Traditionally, such a structure is fixed to be a vector space with data variations represented by translations along individual latent dimensions. We argue this simple structure is suboptimal since it requires the model to learn to discard the properties (e.g. different scales of changes, different levels of abstractness) of data variations, which is an extra work than equivariance learning. Instead, we propose to encode the data variations with groups, a structure not only can equivariantly represent variations, but can also be adaptively optimized to preserve the properties of data variations. Considering it is hard to conduct training on group structures, we focus on Lie groups and adopt a parameterization using Lie algebra. Based on the parameterization, some disentanglement learning constraints are naturally derived. A simple model named Commutative Lie Group VAE is introduced to realize the group-based disentanglement learning. Experiments show that our model can effectively learn disentangled representations without supervision, and can achieve state-of-the-art performance without extra constraints.

For agents

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

get_harvested_code_for_paper("2106.03375")
get_code_for_paper("2106.03375")
have("2106.03375")

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

curl https://syntology.ai/api/ran/2106.03375.json

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

[![Syntology run record](https://syntology.ai/api/ran/2106.03375.svg)](https://syntology.ai/paper/2106.03375)

Connect an agent — have() is free.