SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.14407 · NeurIPS · 2024

Learning an Actionable Discrete Diffusion Policy via Large-Scale Actionless Video Pre-Training

Weinan Zhang, Ling Pan, Chenjia Bai, Bin Zhao, Xuelong Li, Haoran He

arXiv · PDF · Open in the Atlas

Code that ran

17 of 18 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; 1 honoured the contract we drafted, 2 violated it and 4 have no contract verdict; 10 failures of our instrument; 1 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/2402.14407.json · badge: /api/ran/2402.14407.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
peract/peract canonical (classifier label)
classifier basis: census
7 ran · 0 no recorded run
RLAgent/factor-world canonical (classifier label)
classifier basis: census
5 ran · 0 no recorded run
tinnerhrhe/VPDD marked official in Papers with Code
Papers with Code archive
5 ran · 1 no recorded run
FunctionStatusWhere it lives
Huber Ran
ran on a synthesized input · constructed an object; no result was computed
tinnerhrhe/VPDD/diffuser/models/diffusion.py
code served (permissive licence) · get_code("e163906c3cdc2aa6")
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, Huber: 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
from torch import nn
import torch.nn.functional as F

torch.manual_seed(0)

batch_size = 2
horizon = 3
transition_dim = 4
action_dim = 2

weights = torch.ones(horizon, transition_dim)
weights[0, :action_dim] = 2.0

ARGS = {
    'weights': weights,
    'action_dim': action_dim,
}
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 Huber — 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 Huber 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
StateHuber Ran
ran on a synthesized input · constructed an object; no result was computed
tinnerhrhe/VPDD/diffuser/models/diffusion.py
code served (permissive licence) · get_code("acf269bb622e8d6a")
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, StateHuber: 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
from torch import nn
import torch.nn.functional as F

class WeightedStateLoss(nn.Module):

    def __init__(self, weights):
        super().__init__()
        self.register_buffer('weights', weights)

    def forward(self, pred, targ):
        """
            pred, targ : tensor
                [ batch_size x horizon x transition_dim ]
        """
        loss = self._loss(pred, targ)
        weighted_loss = (loss * self.weights).mean()
        a0_loss = (loss[:, 0, :] / self.weights[0, :]).mean()
        return (weighted_loss, {'a0_loss': a0_loss})

class StateHuber(WeightedStateLoss):

    def _loss(self, pred, targ):
        delta = 1.0
        diff = torch.abs(pred - targ)
        cond = diff < delta
        loss = torch.where(cond, F.mse_loss(pred, targ, reduction='none') * 0.5, delta * (torch.abs(pred - targ) - 0.5 * delta))
        return loss

torch.manual_seed(0)

# Create weights tensor: shape [horizon x transition_dim]
# horizon=3, transition_dim=4
weights = torch.ones(3, 4)

# Create the StateHuber model
model = StateHuber(weights=weights)

# Create pred and targ tensors: shape [batch_size x horizon x transition_dim]
# batch_size=2, horizon=3, transition_dim=4
pred = torch.randn(2, 3, 4)
targ = torch.randn(2, 3, 4)

ARGS = {
    'weights': weights,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type StateHuber — 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 StateHuber 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
WeightedStateLoss Ran
ran on a synthesized input · constructed an object; no result was computed
tinnerhrhe/VPDD/diffuser/models/diffusion.py
code served (permissive licence) · get_code("43177911c1fd1d18")
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, WeightedStateLoss: 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)

# The WeightedStateLoss class expects:
# 1. weights: a tensor that will be registered as a buffer
# 2. The forward method calls self._loss(pred, targ) which is not defined in this class,
#    so we need to subclass it and implement _loss

class ConcreteWeightedStateLoss(nn.Module):
    def __init__(self, weights):
        super().__init__()
        self.register_buffer('weights', weights)
    
    def _loss(self, pred, targ):
        # Simple MSE-like loss: [batch_size x horizon x transition_dim]
        return (pred - targ) ** 2
    
    def forward(self, pred, targ):
        """
            pred, targ : tensor
                [ batch_size x horizon x transition_dim ]
        """
        loss = self._loss(pred, targ)
        weighted_loss = (loss * self.weights).mean()
        a0_loss = (loss[:, 0, :] / self.weights[0, :]).mean()
        return (weighted_loss, {'a0_loss': a0_loss})

# Create weights tensor with shape [horizon x transition_dim]
# Using horizon=3, transition_dim=4
weights = torch.ones(3, 4)

ARGS = {
    'weights': weights
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type WeightedStateLoss — 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 WeightedStateLoss 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
act_layer Ran
ran; our contract draft was wrong, not the code
peract/peract/helpers/network_utils.py
code served (permissive licence) · get_code("c99fb7b9654ea364")
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 ReLU. 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)

LRELU_SLOPE = 0.02

def act_layer(act):
    if act == 'relu':
        return nn.ReLU()
    elif act == 'lrelu':
        return nn.LeakyReLU(LRELU_SLOPE)
    elif act == 'elu':
        return nn.ELU()
    elif act == 'tanh':
        return nn.Tanh()
    elif act == 'prelu':
        return nn.PReLU()
    else:
        raise ValueError('%s not recognized.' % act)

ARGS = {
    'act': 'relu'
}
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 ReLU — a ReLU output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ReLU 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
cache_fn Ran
ran; our contract draft was wrong, not the code
peract/peract/agents/peract_bc/perceiver_lang_io.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ThomasMrY/VCT, is pointer only (licence: NONE) · get_code("8458f9ba0361ff43")
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 ThomasMrY/VCT, 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.

default Ran
ran, violated the contract we drafted
peract/peract/agents/peract_bc/perceiver_lang_io.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ThomasMrY/VCT, is pointer only (licence: NONE) · get_code("60fff7c3c400d7ff")
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 ThomasMrY/VCT, 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.

default_sample_fn Ran
ran; our contract draft was wrong, not the code
tinnerhrhe/VPDD/diffuser/models/diffusion.py
code served (permissive licence) · get_code("8d04d499d75080d0")
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 as nn

torch.manual_seed(0)

class MockModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 4)
    
    def p_mean_variance(self, x, cond, task, value, context_mask, t):
        # Returns (model_mean, model_variance, model_log_variance)
        # model_mean should have same shape as x
        model_mean = self.linear(x)
        model_log_variance = torch.zeros_like(x)
        model_variance = torch.exp(model_log_variance)
        return model_mean, model_variance, model_log_variance

# Create tensors with distinct dimensions
batch_size = 2
feature_dim = 4
cond_dim = 3
context_len = 5

model = MockModel()
x = torch.randn(batch_size, feature_dim)
cond = torch.randn(batch_size, cond_dim)
task = torch.randn(batch_size, 2)
value = torch.randn(batch_size, 1)
context_mask = torch.randn(batch_size, context_len)
t = torch.randint(0, 10, (batch_size,))

ARGS = {
    'model': model,
    'x': x,
    'cond': cond,
    'task': task,
    'value': value,
    'context_mask': context_mask,
    't': t,
    'sample_kwargs': {}
}
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
dict_to_list_of_dicts Ran
ran; our contract draft was wrong, not the code
RLAgent/factor-world/envs/factors/factor_wrapper.py
code served (permissive licence) · get_code("051af9d2e2264347")
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 numpy as np

# Create a simple dictionary with string keys and numpy array values
# All arrays must have the same length (first dimension)
d = {
    'a': np.array([[0, 1], [2, 3]]),
    'b': np.array([[4, 5], [6, 7]]),
    'c': np.array([[8, 9], [10, 11]]),
}

ARGS = {
    'd': d,
}
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 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
exists Ran
ran, violated the contract we drafted
peract/peract/agents/peract_bc/perceiver_lang_io.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ThomasMrY/VCT, is pointer only (licence: NONE) · get_code("aa5486a3650902d8")
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 ThomasMrY/VCT, 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.

find_minimum_difference Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
peract/peract/helpers/demo_loading_utils.py
code served (permissive licence) · get_code("0a61ee3eb30e145f")
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

# The function find_minimum_difference takes a list as input
# It accesses lst[-1], iterates through indices, and compares consecutive differences
# It returns a numeric value

# Construct a simple list of numbers
lst = [10, 5, 8, 3, 7, 2]

ARGS = {
    "lst": lst
}
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
full_assets_path_for Ran
ran; our contract draft was wrong, not the code
RLAgent/factor-world/envs/asset_path_utils.py
code served (permissive licence) · get_code("5b55e1bed13f5a58")
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 os
import tempfile
import shutil

# Create a temporary directory to act as the module's directory
temp_module_dir = tempfile.mkdtemp()

# Create an assets subdirectory
assets_dir = os.path.join(temp_module_dir, 'assets')
os.makedirs(assets_dir, exist_ok=True)

# Create a dummy assets file
dummy_file = os.path.join(assets_dir, 'test_file.txt')
with open(dummy_file, 'w') as f:
    f.write('test content')

# Mock the __file__ variable for the function
# We need to monkey-patch the function to use our temp directory
import sys
from types import ModuleType

mock_module = ModuleType('mock_module')
mock_module.__file__ = os.path.join(temp_module_dir, 'mock_module.py')
sys.modules['mock_module'] = mock_module

# Now define the function with the mocked __file__
ASSETS_DIR = os.path.join(os.path.dirname(mock_module.__file__), 'assets')

def full_assets_path_for(filename):
    return os.path.join(ASSETS_DIR, filename)

ARGS = {
    'filename': 'test_file.txt'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
full_mso_path_for Ran
ran; our contract draft was wrong, not the code
RLAgent/factor-world/envs/asset_path_utils.py
code served (permissive licence) · get_code("cda51c68aff53692")
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 os

# The function takes a single parameter: filename (a string)
# It constructs a path by joining MSO_DIR with the filename
# Since MSO_DIR is constructed relative to __file__, we need to mock or work with what we have
# The function simply does string path operations, so any string filename will work

ARGS = {
    "filename": "sample_object.xml"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
full_v2_path_for Ran
ran; our contract draft was wrong, not the code
RLAgent/factor-world/envs/asset_path_utils.py
code served (permissive licence) · get_code("755db90b1e706d42")
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 os

def full_v2_path_for(filename):
    ENV_ASSET_DIR_V2 = os.path.join(os.path.dirname(os.path.dirname(__file__)), 'third_party/metaworld/metaworld/envs/assets_v2')
    return os.path.join(ENV_ASSET_DIR_V2, filename)

ARGS = {
    'filename': 'model.xml'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
norm_layer1d Ran
ran; our contract draft was wrong, not the code
peract/peract/helpers/network_utils.py
code served (permissive licence) · get_code("9ed9933be23b28a5")
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 BatchNorm1d. 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 = {
    'norm': 'batch',
    'num_channels': 64
}
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 BatchNorm1d — a BatchNorm1d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a BatchNorm1d 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
norm_layer2d Ran
ran; our contract draft was wrong, not the code
peract/peract/helpers/network_utils.py
code served (permissive licence) · get_code("d0264b5831943fb9")
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 BatchNorm2d. 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 = {
    'norm': 'batch',
    'channels': 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 BatchNorm2d — a BatchNorm2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a BatchNorm2d 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
sample_without_replacement Ran
ran, honoured the contract we drafted
RLAgent/factor-world/envs/factors/factor_wrapper.py
code served (permissive licence) · get_code("295aa3b49a6888ac")
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 ndarray, ndim 1, dtype family int, shape (5,) 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.

from gym import spaces
import numpy as np

# Test with Discrete space
discrete_space = spaces.Discrete(n=10, start=0)

# Test with Box space
box_space = spaces.Box(low=0.0, high=1.0, shape=(4,), dtype=np.float32)

# Test with Dict space
dict_space = spaces.Dict({
    'discrete_factor': spaces.Discrete(n=5, start=0),
    'box_factor': spaces.Box(low=-1.0, high=1.0, shape=(3,), dtype=np.float32)
})

# Define arguments for different test cases
# Using Discrete space as the primary test
ARGS = {
    'factor_space': discrete_space,
    'n': 5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gym, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family int, shape (5,) — shape (5,) 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
videoloss Ran
ran on a synthesized input · constructed an object; no result was computed
tinnerhrhe/VPDD/diffuser/models/diffusion.py
code served (permissive licence) · get_code("ee809dc4a6634822")
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, videoloss: 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)

# The videoloss class expects:
# - weights: a tensor that will be registered as a buffer
#   Used in forward(): self.weights is multiplied with loss (elementwise)
#   and indexed as self.weights[0, :], so it needs shape [horizon, transition_dim]

batch_size = 2
horizon = 3
transition_dim = 4

# Create weights tensor with shape [horizon, transition_dim]
weights = torch.ones(horizon, transition_dim)

ARGS = {
    'weights': weights,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type videoloss — 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 videoloss 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
DTDiffusion Not yet run
unverified (harvested, no recorded run)
tinnerhrhe/VPDD/diffuser/models/diffusion.py
code served (permissive licence) · get_code("81f1f429fb4a7f3e")
How it ran

no invocation recorded.

Also present in other papers' code

Of 18 samples lifted from this paper's linked repositories, 18 carry a function identity and 0 do not. Of the 18 distinct functions with an identity, 8 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 2 the same month, 6 later) and 10 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
exists 486 other papers
293 dated earlier, 15 the same month, 167 later, 9 same year, order not determinable, 2 undated · 78 only as an assembly brick
1603.08155 (2016-03)
Perceptual Losses for Real-Time Style Transfer and Super-Resolution
1611.07004 (2016-11)
Image-to-Image Translation with Conditional Adversarial Networks
1804.04637 (2018-04)
EMBER: An Open Dataset for Training Static PE Malware Machine Learning Models
and 483 more not named here
default 163 other papers
113 dated earlier, 6 the same month, 41 later, 2 same year, order not determinable, 1 undated · 27 only as an assembly brick
1603.08155 (2016-03)
Perceptual Losses for Real-Time Style Transfer and Super-Resolution
1810.00825 (2018-10)
Set Transformer: A Framework for Attention-based Permutation-Invariant Neural Networks
1907.01470 (2019-07)
Augmenting Self-attention with Persistent Memory
and 160 more not named here
cache_fn 9 other papers
8 dated earlier, 1 later · 2 only as an assembly brick
1810.00825 (2018-10)
Set Transformer: A Framework for Attention-based Permutation-Invariant Neural Networks
2103.03206 (2021-03)
Perceiver: General Perception with Iterative Attention
2107.14795 (2021-07)
Perceiver IO: A General Architecture for Structured Inputs & Outputs
and 6 more not named here
act_layer 5 other papers
4 dated earlier, 1 later
2207.11514 (2022-07)
Semantic Abstraction: Open-World 3D Scene Understanding from 2D Vision-Language Models
2304.04321 (2023-04)
ARNOLD: A Benchmark for Language-Grounded Task Learning With Continuous States in Realistic 3D Scenes
2306.10474 (2023-06)
A Universal Semantic-Geometric Representation for Robotic Manipulation
and 2 more not named here
norm_layer1d 5 other papers
4 dated earlier, 1 later
2207.11514 (2022-07)
Semantic Abstraction: Open-World 3D Scene Understanding from 2D Vision-Language Models
2304.04321 (2023-04)
ARNOLD: A Benchmark for Language-Grounded Task Learning With Continuous States in Realistic 3D Scenes
2306.10474 (2023-06)
A Universal Semantic-Geometric Representation for Robotic Manipulation
and 2 more not named here
norm_layer2d 5 other papers
4 dated earlier, 1 later
2207.11514 (2022-07)
Semantic Abstraction: Open-World 3D Scene Understanding from 2D Vision-Language Models
2304.04321 (2023-04)
ARNOLD: A Benchmark for Language-Grounded Task Learning With Continuous States in Realistic 3D Scenes
2306.10474 (2023-06)
A Universal Semantic-Geometric Representation for Robotic Manipulation
and 2 more not named here
find_minimum_difference 2 other papers
2 dated earlier
2306.10474 (2023-06)
A Universal Semantic-Geometric Representation for Robotic Manipulation
2308.16891 (2023-08)
GNFactor: Multi-Task Real Robot Learning with Generalizable Neural Feature Fields
default_sample_fn
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2305.18459 (2023-05)
Diffusion Model is an Effective Planner and Data Synthesizer for Multi-Task Reinforcement Learning

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

Learning a generalist embodied agent capable of completing multiple tasks poses challenges, primarily stemming from the scarcity of action-labeled robotic datasets. In contrast, a vast amount of human videos exist, capturing intricate tasks and interactions with the physical world. Promising prospects arise for utilizing actionless human videos for pre-training and transferring the knowledge to facilitate robot policy learning through limited robot demonstrations. However, it remains a challenge due to the domain gap between humans and robots. Moreover, it is difficult to extract useful information representing the dynamic world from human videos, because of its noisy and multimodal data structure. In this paper, we introduce a novel framework to tackle these challenges, which leverages a unified discrete diffusion to combine generative pre-training on human videos and policy fine-tuning on a small number of action-labeled robot videos. We start by compressing both human and robot videos into unified video tokens. In the pre-training stage, we employ a discrete diffusion model with a mask-and-replace diffusion strategy to predict future video tokens in the latent space. In the fine-tuning stage, we harness the imagined future videos to guide low-level action learning with a limited set of robot data. Experiments demonstrate that our method generates high-fidelity future videos for planning and enhances the fine-tuned policies compared to previous state-of-the-art approaches with superior performance. Our project webpage is available at https://video-diff.github.io/.

For agents

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

get_harvested_code_for_paper("2402.14407")
get_code_for_paper("2402.14407")
have("2402.14407")

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

curl https://syntology.ai/api/ran/2402.14407.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.14407.svg)](https://syntology.ai/paper/2402.14407)

Connect an agent — have() is free.