SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2006.08205 · NeurIPS · 2020

Learning Latent Space Energy-Based Prior Model

Song-Chun Zhu, Ying Wu, Bo Pang, Tian Han, Erik Nijkamp

arXiv · PDF · Open in the Atlas

Code that ran

6 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (4 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 1 failure of our instrument; 2 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/2006.08205.json · badge: /api/ran/2006.08205.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
bpucla/latent-space-EBM-prior community
Papers with Code archive
6 ran · 2 no recorded run
FunctionStatusWhere it lives
GELU Ran
ran on a synthesized input · constructed an object; no result was computed
bpucla/latent-space-EBM-prior/train_celeba_128.py
code served (permissive licence) · get_code("dcbaf6944f6d645a")
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 an instance of its own class, GELU: the run constructed the object and did not call it, so no result was computed. 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
import torch.nn.functional as F

torch.manual_seed(0)

# GELU has no parameters in __init__, only self which is implicit
# The forward method takes x (a tensor) and applies F.gelu to it

# Create a small test tensor
x = torch.randn(2, 4)

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 GELU — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a GELU 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
Mish Ran
ran on a synthesized input · constructed an object; no result was computed
bpucla/latent-space-EBM-prior/train_celeba_128.py
code served (permissive licence) · get_code("1238b0015976a0a9")
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 an instance of its own class, Mish: the run constructed the object and did not call it, so no result was computed. 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
import torch.nn.functional as F

torch.manual_seed(0)

class Mish(nn.Module):

    def __init__(self):
        super().__init__()

    def forward(self, x):
        return x * torch.tanh(F.softplus(x))


# Create a small test tensor for the forward pass
test_input = torch.randn(2, 4)

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 Mish — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Mish 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
Swish Ran
ran on a synthesized input
bpucla/latent-space-EBM-prior/train_celeba_128.py
code served (permissive licence) · get_code("da8f7cc95b748c33")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) 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
import torch.nn as nn
import torch.nn.functional as F

class Swish(nn.Module):

    def __init__(self):
        super().__init__()

    def forward(self, x):
        return x * F.sigmoid(x)


INIT_ARGS = {}

ARGS = {
    'x': torch.randn(2, 4)
}

if __name__ == '__main__':
    model = Swish(**INIT_ARGS)
    output = model(**ARGS)
    print(output)
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
_netE Ran
ran on a synthesized input · constructed an object; no result was computed
bpucla/latent-space-EBM-prior/train_celeba_128.py
code served (permissive licence) · get_code("8702b4b20711afd9")
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 an instance of its own class, _netE: the run constructed the object and did not call it, so no result was computed. 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.nz = 4
        self.ndf = 8
        self.nez = 3
        self.e_activation = 'gelu'
        self.e_activation_leak = 0.2
        self.e_sn = False

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 _netE — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a _netE 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
_netWrapper Ran
ran on a synthesized input · constructed an object; no result was computed
bpucla/latent-space-EBM-prior/train_celeba_128.py
code served (permissive licence) · get_code("169e5b8fb01ceb7a")
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 an instance of its own class, _netWrapper: the run constructed the object and did not call it, so no result was computed. 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):
        # Network architecture parameters
        self.nz = 4  # latent dimension
        self.ndf = 8  # number of discriminator features
        self.nez = 3  # energy output dimension
        self.ngf = 2  # number of generator features
        self.nc = 1  # number of channels
        
        # Energy network parameters
        self.e_sn = False  # use spectral normalization
        self.e_activation = 'lrelu'  # activation function
        self.e_activation_leak = 0.2  # leaky relu leak
        
        # Generator parameters
        self.g_activation = 'lrelu'
        self.g_batchnorm = True  # use batch norm
        self.gnf = 2  # typo in original code but need to define it
        
        # Langevin prior sampling
        self.e_l_steps = 3  # number of langevin steps
        self.e_l_step_size = 0.1  # step size
        self.e_l_with_noise = True  # add noise during sampling
        self.e_prior_sig = 1.0  # prior sigma
        
        # Langevin posterior sampling
        self.g_l_steps = 3
        self.g_l_step_size = 0.1
        self.g_l_with_noise = True
        self.g_llhd_sigma = 1.0  # likelihood sigma
        
        # Batch size
        self.batch_size = 2

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 _netWrapper — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a _netWrapper 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_activation Ran
ran; our contract draft was wrong, not the code
bpucla/latent-space-EBM-prior/train_celeba_128.py
code served (permissive licence) · get_code("e50f0da40636937b")
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 GELU. 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 MockArgs:
    def __init__(self):
        self.e_activation_leak = 0.01

ARGS = {
    'name': 'gelu',
    'args': MockArgs()
}
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 GELU — a GELU output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a GELU 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
_netG Not yet run
unverified (harvested, no recorded run)
bpucla/latent-space-EBM-prior/train_celeba_128.py
code served (permissive licence) · get_code("28b9ab25eaeb80d5")
How it ran

no invocation recorded.

weights_init_xavier Not yet run
unverified (harvested, no recorded run)
bpucla/latent-space-EBM-prior/train_celeba_128.py
code served (permissive licence) · get_code("74ce9d0a6f6ef663")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 8 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.

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 propose to learn energy-based model (EBM) in the latent space of a generator model, so that the EBM serves as a prior model that stands on the top-down network of the generator model. Both the latent space EBM and the top-down network can be learned jointly by maximum likelihood, which involves short-run MCMC sampling from both the prior and posterior distributions of the latent vector. Due to the low dimensionality of the latent space and the expressiveness of the top-down network, a simple EBM in latent space can capture regularities in the data effectively, and MCMC sampling in latent space is efficient and mixes well. We show that the learned model exhibits strong performances in terms of image and text generation and anomaly detection. The one-page code can be found in supplementary materials.

For agents

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

get_harvested_code_for_paper("2006.08205")
get_code_for_paper("2006.08205")
have("2006.08205")

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

curl https://syntology.ai/api/ran/2006.08205.json

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

[![Syntology run record](https://syntology.ai/api/ran/2006.08205.svg)](https://syntology.ai/paper/2006.08205)

Connect an agent — have() is free.