SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2003.04218 · 2020

Teaching Temporal Logics to Neural Networks

arXiv · PDF · Open in the Atlas

Code that ran

14 of 24 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; 2 honoured the contract we drafted, 0 violated it and 9 have no contract verdict; 3 failures of our instrument; 10 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/2003.04218.json · badge: /api/ran/2003.04218.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
reactive-systems/deepltl canonical (classifier label)
classifier basis: census
5 ran · 8 no recorded run
necrashter/deepltl-pytorch community (classifier label)
classifier basis not recorded on the link
9 ran · 2 no recorded run
FunctionStatusWhere it lives
create_look_ahead_mask Ran
ran, honoured the contract we drafted
reactive-systems/deepltl/deepltl/models/transformer.py
code served (permissive licence) · get_code("1bfc281ca0cb23d6")
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 EagerTensor, ndim 4, dtype family float, shape (1, 1, 3, 3) 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 tensorflow as tf

ARGS = {
    "size": 3,
    "dtype": tf.float32,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 4, dtype family float, shape (1, 1, 3, 3) — shape (1, 1, 3, 3) 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
create_look_ahead_mask Ran
ran on a synthesized input
necrashter/deepltl-pytorch/deepltl/models/transformer.py
code served (permissive licence) · get_code("4bc2b746faf71e08")
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 4, dtype family float, shape (1, 1, 3, 3) 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)

ARGS = {
    "size": 3,
    "device": torch.device("cpu"),
    "dtype": torch.float32,
}
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 4, dtype family float, shape (1, 1, 3, 3) — shape (1, 1, 3, 3) 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
create_padding_mask Ran
ran, honoured the contract we drafted
reactive-systems/deepltl/deepltl/models/transformer.py
code served (permissive licence) · get_code("09cfa39c1a945b25")
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 EagerTensor, ndim 4, dtype family float, shape (2, 1, 1, 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 tensorflow as tf

# Create a small batch of input token IDs with shape (batch_size=2, input_length=4)
# Values range from 0-5, with some padding tokens (pad_id=0)
input_tensor = tf.constant([
    [1, 2, 0, 0],
    [3, 4, 5, 0]
], dtype=tf.int32)

pad_id = 0

dtype = tf.float32

ARGS = {
    "input": input_tensor,
    "pad_id": pad_id,
    "dtype": dtype
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 4, dtype family float, shape (2, 1, 1, 4) — shape (2, 1, 1, 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
create_padding_mask Ran
ran on a synthesized input
necrashter/deepltl-pytorch/deepltl/models/transformer.py
code served (permissive licence) · get_code("f1c292f469811841")
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 4, dtype family float, shape (2, 1, 1, 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)

# Create a small batch of token IDs with padding
batch_size = 2
input_length = 4
pad_id = 0

# Create input tensor with some padding tokens
input_tensor = torch.tensor([
    [1, 2, 0, 0],
    [5, 6, 7, 0]
], dtype=torch.int64)

dtype = torch.float32

ARGS = {
    "input": input_tensor,
    "pad_id": pad_id,
    "dtype": dtype
}
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 4, dtype family float, shape (2, 1, 1, 4) — shape (2, 1, 1, 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
expand_to_same_rank Ran
ran; our contract draft was wrong, not the code
reactive-systems/deepltl/deepltl/models/beam_search.py
code served (permissive licence) · get_code("4d12c68827c9e049")
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 EagerTensor, ndim 4, dtype family float, shape (2, 4, 1, 1) 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 tensorflow as tf

# Create a tensor with rank 2 (2D tensor)
tensor = tf.constant([[1.0, 2.0, 3.0, 4.0],
                       [5.0, 6.0, 7.0, 8.0]])

# Create a target tensor with rank 4 (4D tensor)
target = tf.zeros((2, 3, 4, 5))

ARGS = {
    "tensor": tensor,
    "target": target
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 4, dtype family float, shape (2, 4, 1, 1) — shape (2, 4, 1, 1) 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
flatten_beam_dim Ran
ran on a synthesized input
necrashter/deepltl-pytorch/deepltl/models/beam_search.py
code served (permissive licence) · get_code("b35303fb7949670e")
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 (6, 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 function takes a tensor and flattens the first two dimensions
# It accesses tensor.shape, uses list operations, and calls tensor.view()
# We need a tensor with at least 2 dimensions to demonstrate the flattening

# Create a small tensor with shape (batch=2, beam=3, feature=4)
# This will be flattened to (batch*beam=6, feature=4)
tensor = torch.randn(2, 3, 4)

ARGS = {
    "tensor": tensor
}
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 (6, 4) — shape (6, 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
get_activation Ran
ran on a synthesized input
necrashter/deepltl-pytorch/deepltl/models/transformer.py
code served (permissive licence) · get_code("6959fcf2fd01f0db")
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)

# The function takes a single parameter: activation (str)
# It checks if activation matches certain string values and returns corresponding nn modules
# We need to test with a valid activation name

ARGS = {
    'activation': '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
get_ass Ran
ran on a synthesized input
necrashter/deepltl-pytorch/deepltl/train/common.py
code served (permissive licence) · get_code("c2928e09d7c3cd10")
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)

# The function get_ass takes a list parameter
# It expects the list to have even length
# The list should contain alternating variable names and string values
# Values should be 'True', '1', 'False', or '0'

lst = ['var1', 'True', 'var2', 'False', 'var3', '1', 'var4', '0']

ARGS = {
    'lst': lst
}
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_run_path Ran
ran on a synthesized input
necrashter/deepltl-pytorch/deepltl/train/common.py
code served (permissive licence) · get_code("d6df88fd6fc1d452")
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

ARGS = {
    "dir_name": "results",
    "job_dir": "/tmp/jobs",
    "run_name": "experiment_001"
}
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
get_shape_keep_last_dim Ran
ran; our contract draft was wrong, not the code
reactive-systems/deepltl/deepltl/models/beam_search.py
code served (permissive licence) · get_code("e8a54793bb205026")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned TensorShape. 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 tensorflow as tf
import numpy as np

# Create a small TensorFlow tensor with dynamic shape
# The function expects a tensor and calls get_shape() and tf.shape() on it
tensor = tf.Variable(
    tf.random.normal([2, 4, 3]),
    trainable=False
)

ARGS = {
    "tensor": tensor
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type TensorShape — a TensorShape output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a TensorShape 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
map_structure Ran
ran on a synthesized input
necrashter/deepltl-pytorch/deepltl/models/beam_search.py
code served (permissive licence) · get_code("fba9be9ba7c1e0e0")
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)

# Define a simple function to apply to leaf values
def simple_func(x):
    if isinstance(x, torch.Tensor):
        return x * 2
    elif isinstance(x, (int, float)):
        return x + 1
    else:
        return x

# Create a complex nested structure with dicts, lists, tuples, and leaf values
nested = {
    "tensor_leaf": torch.randn(2, 4),
    "int_leaf": 5,
    "list_of_tensors": [torch.randn(3, 4), torch.randn(2, 4)],
    "tuple_of_values": (torch.randn(2, 4), 10, 3.14),
    "nested_dict": {
        "inner_tensor": torch.randn(2, 4),
        "inner_list": [1, 2, torch.randn(2, 4)],
        "inner_tuple": (torch.randn(2, 4), "text", 42)
    }
}

ARGS = {
    "func": simple_func,
    "nested": nested
}
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
positional_encoding Ran
ran on a synthesized input
necrashter/deepltl-pytorch/deepltl/layers/positional_encoding.py
code served (permissive licence) · get_code("f3d335e856a3e9e4")
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 3, dtype family float, shape (1, 5, 8) 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 numpy as np
import torch

torch.manual_seed(0)

ARGS = {
    "position": 5,
    "d_embedding": 8
}
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 Tensor, ndim 3, dtype family float, shape (1, 5, 8) — shape (1, 5, 8) 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
scaled_dot_product_attention Ran
ran on a synthesized input
necrashter/deepltl-pytorch/deepltl/layers/attention.py
code served (permissive licence) · get_code("73b90c22738f0032")
How it ran

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

import torch
import torch.nn.functional as F

torch.manual_seed(0)

# Define dimensions to be distinct so shapes can be probed
batch_size = 2
num_queries = 3
num_keys = 4
d_queries = 5
d_values = 6

# Create queries: (batch_size, num_queries, d_queries)
queries = torch.randn(batch_size, num_queries, d_queries)

# Create keys: (batch_size, num_keys, d_queries)
keys = torch.randn(batch_size, num_keys, d_queries)

# Create values: (batch_size, num_keys, d_values)
values = torch.randn(batch_size, num_keys, d_values)

# Create mask: (batch_size, num_queries, num_keys)
# Use a lower triangular mask as an example (causal mask)
mask = torch.triu(torch.ones(num_queries, num_keys) * float('-inf'), diagonal=1)
mask = mask.unsqueeze(0).expand(batch_size, -1, -1)

ARGS = {
    "queries": queries,
    "keys": keys,
    "values": values,
    "mask": mask,
}
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
shape_list Ran
ran; our contract draft was wrong, not the code
reactive-systems/deepltl/deepltl/models/beam_search.py
code served (permissive licence) · get_code("e092e8e3a00bcec6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned list, ndim 1, dtype family int; 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 tensorflow as tf

# Create a TensorFlow tensor with both static and dynamic dimensions
# The function calls get_shape().as_list() and tf.shape()
# We'll create a tensor with a None dimension to test the dynamic shape handling

# Create a tensor with a partially known shape
tensor = tf.keras.Input(shape=(4,), batch_size=None)

# Alternatively, create a concrete tensor with known shape
tensor = tf.constant([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]])

ARGS = {
    "tensor": tensor
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
tensorflow — 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 int — 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
abbrev_count Not yet run
unverified (harvested, no recorded run)
reactive-systems/deepltl/deepltl/utils/utils.py
code served (permissive licence) · get_code("16bdc764a81d1f5a")
How it ran

no invocation recorded.

checkpoint_path Not yet run
unverified (harvested, no recorded run)
reactive-systems/deepltl/deepltl/train/common.py
code served (permissive licence) · get_code("bd594cfa2fd652e7")
How it ran

no invocation recorded.

dataset_name Not yet run
unverified (harvested, no recorded run)
reactive-systems/deepltl/deepltl/utils/utils.py
code served (permissive licence) · get_code("3e033cf884fa1ed7")
How it ran

no invocation recorded.

get_angles Not yet run
unverified (harvested, no recorded run)
reactive-systems/deepltl/deepltl/layers/positional_encoding.py
code served (permissive licence) · get_code("370c1f3fe2690649")
How it ran

no invocation recorded.

get_latest_checkpoint Not yet run
unverified (harvested, no recorded run)
necrashter/deepltl-pytorch/deepltl/train/common.py
code served (permissive licence) · get_code("981652cdff73c2e9")
How it ran

no invocation recorded.

get_shape_keep_last_dim Not yet run
unverified (harvested, no recorded run)
necrashter/deepltl-pytorch/deepltl/models/beam_search.py
code served (permissive licence) · get_code("0fb71688c64be6e5")
How it ran

no invocation recorded.

positional_encoding Not yet run
unverified (harvested, no recorded run)
reactive-systems/deepltl/deepltl/layers/positional_encoding.py
code served (permissive licence) · get_code("ab5aea947fca2ea9")
How it ran

no invocation recorded.

prepare_dataset_no_tf Not yet run
unverified (harvested, no recorded run)
reactive-systems/deepltl/deepltl/train/common.py
code served (permissive licence) · get_code("2f3557097c940388")
How it ran

no invocation recorded.

run_with_timeout Not yet run
unverified (harvested, no recorded run)
reactive-systems/deepltl/deepltl/utils/utils.py
code served (permissive licence) · get_code("817e786e14f0e285")
How it ran

no invocation recorded.

scaled_dot_product_attention Not yet run
unverified (harvested, no recorded run)
reactive-systems/deepltl/deepltl/layers/attention.py
code served (permissive licence) · get_code("6ac8c2e8189b5ab7")
How it ran

no invocation recorded.

Also present in other papers' code

Of 24 samples lifted from this paper's linked repositories, 24 carry a function identity and 0 do not. Of the 24 distinct functions with an identity, 7 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 7 later) and 17 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
create_look_ahead_mask 3 other papers
1 dated earlier, 2 later · 1 only as an assembly brick
1706.03762 (2017-06)
Attention Is All You Need
2006.09265 (2020-06)
ISARSTEP: A BENCHMARK FOR HIGH-LEVEL MATHEMATICAL REASONING
2110.10054 (2021-10)
Generating Symbolic Reasoning Problems with Transformer GANs
create_padding_mask 2 other papers
1 dated earlier, 1 later · 1 only as an assembly brick
1706.03762 (2017-06)
Attention Is All You Need
2006.09265 (2020-06)
ISARSTEP: A BENCHMARK FOR HIGH-LEVEL MATHEMATICAL REASONING
expand_to_same_rank 2 other papers
1 dated earlier, 1 later · 1 only as an assembly brick
1706.03762 (2017-06)
Attention Is All You Need
2006.09265 (2020-06)
ISARSTEP: A BENCHMARK FOR HIGH-LEVEL MATHEMATICAL REASONING
get_shape_keep_last_dim 2 other papers
1 dated earlier, 1 later · 1 only as an assembly brick
1706.03762 (2017-06)
Attention Is All You Need
2006.09265 (2020-06)
ISARSTEP: A BENCHMARK FOR HIGH-LEVEL MATHEMATICAL REASONING
shape_list 2 other papers
1 dated earlier, 1 later · 1 only as an assembly brick
1706.03762 (2017-06)
Attention Is All You Need
2006.09265 (2020-06)
ISARSTEP: A BENCHMARK FOR HIGH-LEVEL MATHEMATICAL REASONING
checkpoint_path 1 other paper
1 later
2110.10054 (2021-10)
Generating Symbolic Reasoning Problems with Transformer GANs
get_angles 1 other paper
1 later
2110.10054 (2021-10)
Generating Symbolic Reasoning Problems with Transformer GANs

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 study two fundamental questions in neuro-symbolic computing: can deep learning tackle challenging problems in logics end-to-end, and can neural networks learn the semantics of logics. In this work we focus on linear-time temporal logic (LTL), as it is widely used in verification. We train a Transformer on the problem to directly predict a solution, i.e. a trace, to a given LTL formula. The training data is generated with classical solvers, which, however, only provide one of many possible solutions to each formula. We demonstrate that it is sufficient to train on those particular solutions to formulas, and that Transformers can predict solutions even to formulas from benchmarks from the literature on which the classical solver timed out. Transformers also generalize to the semantics of the logics: while they often deviate from the solutions found by the classical solvers, they still predict correct solutions to most formulas.

For agents

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

get_harvested_code_for_paper("2003.04218")
get_code_for_paper("2003.04218")
have("2003.04218")

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

curl https://syntology.ai/api/ran/2003.04218.json

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

[![Syntology run record](https://syntology.ai/api/ran/2003.04218.svg)](https://syntology.ai/paper/2003.04218)

Connect an agent — have() is free.