SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1807.06756 · 2018

SySeVR: A Framework for Using Deep Learning to Detect Software Vulnerabilities

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 (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 5 have no contract verdict; 0 failures of our instrument; 2 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, 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/1807.06756.json · badge: /api/ran/1807.06756.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
SySeVR/SySeVR canonical (classifier label)
classifier basis: classifier_2025
0 ran · 2 no recorded run
NIPC-DL/covec community (classifier label)
classifier basis not recorded on the link
5 ran · 0 no recorded run
copy not recorded community
not in the paper's declared repositories
1 ran · 0 no recorded run
FunctionStatusWhere it lives
is_number Ran
ran, violated the contract we drafted
this paper's copy was not recorded; identical code first harvested from meta-math/MetaMath
pointer only · get_code("0fe9cf6c00ef56ea")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

code_split Ran
ran on a synthesized input
NIPC-DL/covec/torchplp/processor/textmodel.py
code served (permissive licence) · get_code("2bbacb87c24f2c33")
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 re

def code_split(line):
    """split code line into atom
    
    Args:
        line (str): The code line
    
    Return:
        tokens (list): The list of atom token for code line

    """
    return list(filter(lambda x: x and x not in [' ', ''], re.split('(\\W|\\s)', line)))


ARGS = {
    "line": "x = y + 1"
}
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 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
l2o Ran
ran on a synthesized input
NIPC-DL/covec/torchplp/datasets/models.py
code served (permissive licence) · get_code("208a5dff58ff3b6c")
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 (1, 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)

# The function l2o takes y as input and uses:
# - torch.FloatTensor(1, 2) to create a tensor
# - y.zero_() to zero it out
# - y.scatter_(1, y, 1) to scatter values
# The scatter_ call expects y to be a LongTensor with indices

# y should be a LongTensor with shape that can be scattered into dimension 1
# of a (1, 2) tensor. For scatter_ with dim=1, y should have shape (1, k)
# where k is the number of indices to scatter.

y = torch.LongTensor([[0]])

ARGS = {
    "y": y
}
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 (1, 2) — shape (1, 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
standardize Ran
ran on a synthesized input
NIPC-DL/covec/torchplp/processor/treemodel.py
code served (permissive licence) · get_code("6e760cff468b1a28")
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 Node. 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.

class Node:
    def __init__(self, kind=None, data=None):
        self.kind = kind
        self.data = data
        self.children = []
    
    def walk(self):
        """Generator that yields this node and all descendants"""
        yield self
        for child in self.children:
            yield from child.walk()
    
    def add_child(self, child):
        self.children.append(child)


# Construct a simple AST tree for testing
root = Node(kind='PROGRAM', data='root')

# Add a variable declaration node
var_decl_1 = Node(kind='VAR_DECL', data='x')
root.add_child(var_decl_1)

# Add a variable declaration node
var_decl_2 = Node(kind='VAR_DECL', data='y')
root.add_child(var_decl_2)

# Add a function node with CWE in the name
func_node_1 = Node(kind='FUNC_CALL', data='CWE_123_buffer_overflow')
root.add_child(func_node_1)

# Add another function node with CWE
func_node_2 = Node(kind='FUNC_CALL', data='CWE_456_use_after_free')
root.add_child(func_node_2)

# Add references to variables and functions in the tree
var_ref_1 = Node(kind='VAR_REF', data='x')
func_node_1.add_child(var_ref_1)

var_ref_2 = Node(kind='VAR_REF', data='y')
func_node_2.add_child(var_ref_2)

func_ref = Node(kind='FUNC_REF', data='CWE_123_buffer_overflow')
func_node_1.add_child(func_ref)

ARGS = {
    'root': root
}
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 Node — a Node output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Node 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
vectorlize Ran
ran on a synthesized input
NIPC-DL/covec/torchplp/processor/textmodel.py
code served (permissive licence) · get_code("477594548fdbff4e")
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 2, dtype family float, shape (5, 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 numpy as np
import torch

# Create a small embedder (dict-like object that maps words to vector embeddings)
# The function accesses embedder[word] and calls .tolist() on the result
embedder = {
    'def': np.array([0.1, 0.2, 0.3, 0.4]),
    'function': np.array([0.5, 0.6, 0.7, 0.8]),
    'return': np.array([0.9, 1.0, 1.1, 1.2]),
    'x': np.array([1.3, 1.4, 1.5, 1.6]),
    'y': np.array([1.7, 1.8, 1.9, 2.0]),
}

# Create symbolic representation: list of lists of words
# Each inner list represents a line of code
sr = [
    ['def', 'function'],
    ['return', 'x'],
    ['y']
]

ARGS = {
    'sr': sr,
    'embedder': embedder,
}
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 ndarray, ndim 2, dtype family float, shape (5, 4) — shape (5, 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
vectorlize Ran
ran on a synthesized input
NIPC-DL/covec/torchplp/processor/treemodel.py
code served (permissive licence) · get_code("4039685e38853fdd")
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 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 numpy as np

class MockEmbedder:
    """Stand-in for an embedder that supports indexing and has vector_size attribute"""
    def __init__(self, vector_size=4):
        self.vector_size = vector_size
        # Create a dict-like embedding space with a few vectors
        self._embeddings = {
            'node_a': np.array([1.0, 2.0, 3.0, 4.0]),
            'node_b': np.array([5.0, 6.0, 7.0, 8.0]),
            'node_c': np.array([9.0, 10.0, 11.0, 12.0]),
        }
    
    def __getitem__(self, key):
        if key in self._embeddings:
            return self._embeddings[key]
        raise KeyError(f"Key {key} not found")

# Create sample data: list of tuples where each tuple has two node identifiers
data = [
    ('node_a', 'node_b'),
    ('node_c', 'node_a'),
]

# Create embedder with vector_size attribute
embedder = MockEmbedder(vector_size=4)

ARGS = {
    'data': data,
    'embedder': embedder,
}
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 ndarray, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
model_from_config Not yet run
unverified (harvested, no recorded run)
SySeVR/SySeVR/Implementation/model/keras/models.py
code served (permissive licence) · get_code("421eed064e01b9e4")
How it ran

no invocation recorded.

save_model Not yet run
unverified (harvested, no recorded run)
SySeVR/SySeVR/Implementation/model/keras/models.py
code served (permissive licence) · get_code("757b53c434bfd276")
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, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 7 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
is_number 18 other papers
18 later
1906.02425 (2019-06)
Uncertainty-guided Continual Learning with Bayesian Neural Networks
2005.00558 (2020-05)
POINTER: Constrained Progressive Text Generation via Insertion-based Generative Pre-training
2201.06776 (2022-01)
Pruning-aware Sparse Regularization for Network Pruning
and 15 more not named here

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

The detection of software vulnerabilities (or vulnerabilities for short) is an important problem that has yet to be tackled, as manifested by the many vulnerabilities reported on a daily basis. This calls for machine learning methods for vulnerability detection. Deep learning is attractive for this purpose because it alleviates the requirement to manually define features. Despite the tremendous success of deep learning in other application domains, its applicability to vulnerability detection is not systematically understood. In order to fill this void, we propose the first systematic framework for using deep learning to detect vulnerabilities in C/C++ programs with source code. The framework, dubbed Syntax-based, Semantics-based, and Vector Representations (SySeVR), focuses on obtaining program representations that can accommodate syntax and semantic information pertinent to vulnerabilities. Our experiments with 4 software products demonstrate the usefulness of the framework: we detect 15 vulnerabilities that are not reported in the National Vulnerability Database. Among these 15 vulnerabilities, 7 are unknown and have been reported to the vendors, and the other 8 have been "silently" patched by the vendors when releasing newer versions of the pertinent software products.

For agents

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

get_harvested_code_for_paper("1807.06756")
get_code_for_paper("1807.06756")
have("1807.06756")

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

curl https://syntology.ai/api/ran/1807.06756.json

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

[![Syntology run record](https://syntology.ai/api/ran/1807.06756.svg)](https://syntology.ai/paper/1807.06756)

Connect an agent — have() is free.