SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.01850 · 2020

Variational Autoencoders for Anomalous Jet Tagging

arXiv · PDF · Open in the Atlas

Code that ran

8 of 9 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 7 have no contract verdict; 0 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/2007.01850.json · badge: /api/ran/2007.01850.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
taolicheng/VAE-Jet canonical (classifier label)
classifier basis: census
8 ran · 1 no recorded run
FunctionStatusWhere it lives
annealing_fn Ran
ran on a synthesized input
taolicheng/VAE-Jet/tf2/train_disco.py
code served (permissive licence) · get_code("9c9155005989a5a6")
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 float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function takes two parameters: epoch (int) and weight (float)
# It performs annealing logic based on epoch number and returns updated weight

epoch = 50
weight = 0.5

ARGS = {
    "epoch": epoch,
    "weight": weight,
}
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 float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
annealing_fn Ran
ran on a synthesized input
taolicheng/VAE-Jet/tf2/train_oe_resample.py
code served (permissive licence) · get_code("d361c031d072ede4")
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 float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# The function annealing_fn takes two parameters:
# - epoch: an integer representing the current training epoch
# - weight: a float representing the current KL weight

ARGS = {
    "epoch": 0,
    "weight": 0.0
}
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 float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
annealing_fn_2steps Ran
ran on a synthesized input
taolicheng/VAE-Jet/tf2/train_disco.py
code served (permissive licence) · get_code("ec98a116d9f61d78")
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 float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

ARGS = {
    "epoch": 15,
    "weight": 0.5
}
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 float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
annealing_fn_v2 Ran
ran on a synthesized input
taolicheng/VAE-Jet/tf2/train_oe_resample.py
code served (permissive licence) · get_code("6a6ad31f7157364e")
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 float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

# Set seed for reproducibility
torch.manual_seed(0)

# The function annealing_fn_v2 takes:
# - epoch: an integer representing the current epoch
# - weight: a float representing the current weight value

# We'll test with a few different epoch values to show the function works
# Let's create arguments for a representative case

ARGS = {
    "epoch": 15,
    "weight": 0.5
}
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 float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
distance_corr Ran
ran on a synthesized input
taolicheng/VAE-Jet/tf2/disco.py
code served (permissive licence) · get_code("824ab9ee98133ee2")
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 SymbolicTensor. 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

# Set seed for reproducibility
np.random.seed(0)
tf.random.set_seed(0)

# Create small 1D tensors with the same number of entries
n_samples = 4

# var_1: First variable to decorrelate (e.g., mass) - 1D tensor
var_1 = tf.constant([1.0, 2.0, 3.0, 4.0], dtype=tf.float32)

# var_2: Second variable to decorrelate (e.g., classifier output) - 1D tensor
var_2 = tf.constant([0.1, 0.5, 0.8, 0.9], dtype=tf.float32)

# normedweight: Per-example weight normalized so sum equals N
# Sum should equal n_samples (4)
normedweight = tf.constant([1.0, 1.0, 1.0, 1.0], dtype=tf.float32)

# power: Exponent used in calculating distance correlation
power = 1

ARGS = {
    "var_1": var_1,
    "var_2": var_2,
    "normedweight": normedweight,
    "power": power,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SymbolicTensor — a SymbolicTensor 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
jet_mass Ran
ran, violated the contract we drafted
taolicheng/VAE-Jet/tf1/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from taolicheng/ebm-hep, is pointer only (licence: NONE) · get_code("55fa93b1134abf90")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from taolicheng/ebm-hep, 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.

jet_pt Ran
ran on a synthesized input
taolicheng/VAE-Jet/tf1/utils.py
code served (permissive licence) · get_code("e28433fdc4f312cc")
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 float32, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# The function takes a 'jet' parameter which should be reshaped to (-1, 4)
# It iterates through constituents and accesses columns 1 and 2
# So we need an array-like object with shape compatible with (-1, 4)

# Create a simple jet array with 3 constituents, each with 4 components
# Components appear to be: [E, Px, Py, Pz] based on usage (indices 1 and 2 for momentum)
jet_data = np.array([
    [10.0, 2.0, 3.0, 1.0],  # constituent 1
    [8.0, 1.5, 2.5, 0.8],   # constituent 2
    [6.0, 0.5, 1.0, 0.3],   # constituent 3
], dtype=np.float32)

ARGS = {
    'jet': jet_data
}
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 float32, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_scaler Ran
ran on a synthesized input
taolicheng/VAE-Jet/tf1/utils.py
code served (permissive licence) · get_code("5c254e62e9e1ba97")
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 RobustScaler. 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 h5py
import numpy as np
import tempfile
import os
from sklearn.preprocessing import RobustScaler

# Create a temporary HDF5 file with sample data
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, 'test_data.h5')

# Generate synthetic jet data: shape (10, 80) for 10 jets with 80 features
# Each jet has 20 constituents × 4 features (E, Px, Py, Pz)
np.random.seed(42)
synthetic_jet_data = np.random.randn(10, 80).astype(np.float32)

# Create the HDF5 file with the 'table' key
with h5py.File(temp_file, 'w') as f:
    f.create_dataset('table', data=synthetic_jet_data)

ARGS = {
    'filename': temp_file,
    'key': 'table',
    'len_input': 80,
    'pt_scaling': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
h5py, numpy, sklearn — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type RobustScaler — a RobustScaler output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a RobustScaler 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
load_test Not yet run
unverified (harvested, no recorded run)
taolicheng/VAE-Jet/tf2/utils.py
code served (permissive licence) · get_code("aac9738b35e2457f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 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 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.

FunctionAlso present inEarliest-datedOthers named
jet_mass 1 other paper
1 later
2302.00695 (2023-02)
Versatile Energy-Based Probabilistic Models for High Energy Physics

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 present a detailed study on Variational Autoencoders (VAEs) for anomalous jet tagging at the Large Hadron Collider. By taking in low-level jet constituents' information, and training with background QCD jets in an unsupervised manner, the VAE is able to encode important information for reconstructing jets, while learning an expressive posterior distribution in the latent space. When using the VAE as an anomaly detector, we present different approaches to detect anomalies: directly comparing in the input space or, instead, working in the latent space. In order to facilitate general search approaches such as bump-hunt, mass-decorrelated VAEs based on distance correlation regularization are also studied. We find that the naive mass-decorrelated VAEs fail at maintaining proper detection performance, by assigning higher probabilities to some anomalous samples. To build a performant mass-decorrelated anomalous jet tagger, we propose the Outlier Exposed VAE (OE-VAE), for which some outlier samples are introduced in the training process to guide the learned information. OE-VAEs are employed to achieve two goals at the same time: increasing sensitivity of outlier detection and decorrelating jet mass from the anomaly score. We succeed in reaching excellent results from both aspects. Code implementation of this work can be found at https://github.com/taolicheng/VAE-Jet

For agents

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

get_harvested_code_for_paper("2007.01850")
get_code_for_paper("2007.01850")
have("2007.01850")

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

curl https://syntology.ai/api/ran/2007.01850.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.01850.svg)](https://syntology.ai/paper/2007.01850)

Connect an agent — have() is free.