SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.20366 · NeurIPS · 2024

Rethinking Reconstruction-based Graph-Level Anomaly Detection: Limitations and a Simple Remedy

Shinhwan Kang, Kijung Shin, Sunwoo Kim, Soo Lee, Fanchen Bu, Jaemin Yoo, Kyungho Kim

arXiv · PDF · Open in the Atlas

Code that ran

8 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 6 failures of our instrument; 0 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/2410.20366.json · badge: /api/ran/2410.20366.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
copy not recorded community
not in the paper's declared repositories
3 ran · 0 no recorded run
huggingface/open-muse reimplementation (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
kswoo97/GLAD_MUSE marked official in Papers with Code
Papers with Code archive
2 ran · 0 no recorded run
FunctionStatusWhere it lives
MUSE_representation_learning Ran
ran on a synthesized input · constructed an object; no result was computed
kswoo97/GLAD_MUSE/src.py
pointer only (licence: NONE) · get_code("6b3abe06c8b61a63")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kswoo97/GLAD_MUSE 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.

SubsetSampler Ran
ran on a synthesized input · constructed an object; no result was computed
kswoo97/GLAD_MUSE/src.py
pointer only (licence: NONE) · get_code("c25a2af8838a5048")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kswoo97/GLAD_MUSE 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.

make_attention_mask Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from huggingface/open-muse
pointer only · get_code("b22b48bb429101bb")
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.

make_attention_mask Ran
ran; our contract draft was wrong, not the code
huggingface/open-muse/muse/modeling_transformer.py
code served (permissive licence) · get_code("e88526ab8fc3ebc4")
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 Tensor, ndim 4, dtype family float, shape (2, 1, 3, 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 small tensors for query_input and key_input
# query_input shape: (batch=2, query_len=3)
query_input = torch.tensor([[1.0, 0.0, 1.0], [0.0, 1.0, 0.0]], dtype=torch.float32)

# key_input shape: (batch=2, key_len=4)
key_input = torch.tensor([[1.0, 1.0, 0.0, 1.0], [0.0, 1.0, 1.0, 0.0]], dtype=torch.float32)

# pairwise_fn: a callable that takes two tensors and returns element-wise product
pairwise_fn = torch.mul

ARGS = {
    "query_input": query_input,
    "key_input": key_input,
    "pairwise_fn": pairwise_fn,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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, 3, 4) — shape (2, 1, 3, 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
prob_mask_like Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
this paper's copy was not recorded; identical code first harvested from huggingface/open-muse
pointer only · get_code("1a6c10e37cb97711")
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.

prob_mask_like Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
huggingface/open-muse/muse/modeling_transformer.py
code served (permissive licence) · get_code("162f6eae613812c6")
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 Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

ARGS = {
    "shape": (2, 4),
    "prob": 0.5,
    "device": None,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
uniform Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from huggingface/open-muse
pointer only · get_code("7daf2f4e6899de97")
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.

uniform Ran
ran; our contract draft was wrong, not the code
huggingface/open-muse/muse/modeling_transformer.py
code served (permissive licence) · get_code("66e51521e0875290")
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 Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

ARGS = {
    "shape": (2, 4),
    "min": 0,
    "max": 1,
    "device": None,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

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 5 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 3 the same month, 0 later) and 2 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
make_attention_mask 2 other papers
1 dated earlier, 1 the same month
2301.00704 (2023-01)
Muse: Text-To-Image Generation via Masked Generative Transformers
2410.13136 (2024-10)
Unlocking the Capabilities of Masked Generative Models for Image Synthesis via Self-Guidance
prob_mask_like 2 other papers
1 dated earlier, 1 the same month
2301.00704 (2023-01)
Muse: Text-To-Image Generation via Masked Generative Transformers
2410.13136 (2024-10)
Unlocking the Capabilities of Masked Generative Models for Image Synthesis via Self-Guidance
uniform 2 other papers
1 dated earlier, 1 the same month
2301.00704 (2023-01)
Muse: Text-To-Image Generation via Masked Generative Transformers
2410.13136 (2024-10)
Unlocking the Capabilities of Masked Generative Models for Image Synthesis via Self-Guidance

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

Graph autoencoders (Graph-AEs) learn representations of given graphs by aiming to accurately reconstruct them. A notable application of Graph-AEs is graph-level anomaly detection (GLAD), whose objective is to identify graphs with anomalous topological structures and/or node features compared to the majority of the graph population. Graph-AEs for GLAD regard a graph with a high mean reconstruction error (i.e. mean of errors from all node pairs and/or nodes) as anomalies. Namely, the methods rest on the assumption that they would better reconstruct graphs with similar characteristics to the majority. We, however, report non-trivial counterexamples, a phenomenon we call reconstruction flip, and highlight the limitations of the existing Graph-AE-based GLAD methods. Specifically, we empirically and theoretically investigate when this assumption holds and when it fails. Through our analyses, we further argue that, while the reconstruction errors for a given graph are effective features for GLAD, leveraging the multifaceted summaries of the reconstruction errors, beyond just mean, can further strengthen the features. Thus, we propose a novel and simple GLAD method, named MUSE. The key innovation of MUSE involves taking multifaceted summaries of reconstruction errors as graph features for GLAD. This surprisingly simple method obtains SOTA performance in GLAD, performing best overall among 14 methods across 10 datasets. This counterintuitive phenomenon guides our analysis, implication, and the proposed method. First, we aim to understand this counterintuitive phenomenon, which we refer to as the reconstruction flip. 38th Conference on Neural Information Processing Systems (NeurIPS 2024).

For agents

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

get_harvested_code_for_paper("2410.20366")
get_code_for_paper("2410.20366")
have("2410.20366")

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

curl https://syntology.ai/api/ran/2410.20366.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.20366.svg)](https://syntology.ai/paper/2410.20366)

Connect an agent — have() is free.