SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2404.02115 · NAACL · 2024

GINopic: Topic Modeling with Graph Isomorphism Network

Suman Adhya, Debarshi Sanyal

arXiv · PDF · Open in the Atlas

Code that ran

10 of 12 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; 1 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 8 failures of our instrument; 2 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/2404.02115.json · badge: /api/ran/2404.02115.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
SmilesDZgk/GNTM canonical (classifier label)
classifier basis: census
7 ran · 1 no recorded run
BobXWu/ECRTM canonical (classifier label)
classifier basis: census
3 ran · 0 no recorded run
valdersoul/GraphBTM community
Papers with Code archive
0 ran · 1 no recorded run
FunctionStatusWhere it lives
Expectation_log_Dirichlet Ran
ran; our contract draft was wrong, not the code
SmilesDZgk/GNTM/modules/GDGNNmodel.py
code served (permissive licence) · get_code("1fddcdcf09bd6e79")
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 (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)

# gamma should be a tensor that works with digamma and sum operations
# The function computes digamma(gamma) - digamma(gamma.sum(dim=-1, keepdim=True))
# So gamma should be a multi-dimensional tensor where the last dimension represents
# the Dirichlet parameters

gamma = torch.randn(2, 4) + 2.0  # Add offset to ensure positive values for digamma stability
gamma = torch.abs(gamma) + 1.0   # Ensure strictly positive for digamma

ARGS = {
    "gamma": gamma
}
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 (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
L2_norm Ran
ran; our contract draft was wrong, not the code
SmilesDZgk/GNTM/modules/GDGNNmodel.py
code served (permissive licence) · get_code("b35e5d98a6ff271b")
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 (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)

# Create a small tensor for testing L2_norm
# The function expects a tensor and normalizes along the last dimension
# Shape: (batch_size=2, features=4)
vec = torch.randn(2, 4)

ARGS = {
    "vec": vec
}
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 (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
NMI Ran
ran; our contract draft was wrong, not the code
SmilesDZgk/GNTM/utils.py
code served (permissive licence) · get_code("99350b1b0466803a")
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 float64, ndim 0, 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 simple test data for NMI function
# labels: ground truth cluster assignments
# preds: predicted cluster assignments

np.random.seed(0)

# Create labels: 10 samples with 3 true clusters (0, 1, 2)
labels = np.array([0, 0, 0, 1, 1, 1, 2, 2, 2, 2], dtype=np.int64)

# Create preds: predicted clusters (should be somewhat similar to labels)
# Using 3 clusters as well
preds = np.array([0, 0, 1, 1, 1, 2, 2, 2, 0, 2], dtype=np.int64)

ARGS = {
    "labels": labels,
    "preds": preds,
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
TD_eva Ran
ran, honoured the contract we drafted
BobXWu/ECRTM/ECRTM/utils/eva/TD.py
pointer only (licence: NONE) · get_code("d4d25942e0f00f0e")
How it ran

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

clean_words Ran
ran; our contract draft was wrong, not the code
SmilesDZgk/GNTM/dataPrepare/preprocess.py
code served (permissive licence) · get_code("8cf0e3270ba0dfd9")
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

stopwords = {"max>'ax>'ax>'ax>'ax>'ax>'ax>'ax>'ax>'ax>'ax>'ax>'ax>'ax>'ax": 1, 'edu': 1, 'subject': 1, 'com': 1, 'r<g': 1, '_?w': 1, 'isc': 1, 'cx^': 1, 'usr': 1, 'uga': 1, 'sam': 1, 'mhz': 1, 'b8f': 1, '34u': 1, 'pl+': 1, '1993apr20': 1, '1993apr15': 1, 'xterm': 1, 'utexas': 1, 'x11r5': 1, 'o+r': 1, 'iastate': 1, 'udel': 1, 'uchicago': 1, '1993apr21': 1, 'uxa': 1, 'argic': 1, 'optilink': 1, 'imho': 1, 'umich': 1, 'openwindows': 1, '1993apr19': 1, '1993apr22': 1, 'unk': 1}

word_len_threshold = 2

def is_num(w):
    try:
        float(w)
        return True
    except:
        False

def clean_words(words):
    new_words = []
    for word in words:
        word = word.lower()
        if word in stopwords:
            continue
        word = word.strip('_\\[\\]\'".,()*! #@~`\\%^&;:/-+=""''<>{}|?$^&').replace("isn't", '').replace("'s", '').replace("'re", '').replace("'t", '').replace("'ll", '').replace("'m", '').replace("'am", '').replace("'ve", '').replace("'d", '')
        segs = re.split('[()@.\\-/#\\\\"`\\[\\]=:&<>%\']', word)
        new_word = []
        for s in segs:
            seg = s
            if seg not in stopwords and seg and (len(seg) > word_len_threshold) and (not is_num(seg)):
                new_word.append(seg)
        new_words.extend(new_word)
    return new_words

ARGS = {
    'words': ['hello', 'world', 'test', 'python', 'code', 'example', 'machine', 'learning']
}
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
is_num Ran
ran, violated the contract we drafted
SmilesDZgk/GNTM/dataPrepare/preprocess.py
code served (permissive licence) · get_code("ade702289ccc9871")
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 bool, 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

def is_num(w):
    try:
        float(w)
        return True
    except:
        False

ARGS = {
    "w": "42.5"
}
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 bool, 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
parse_sent Ran
ran; our contract draft was wrong, not the code
SmilesDZgk/GNTM/dataPrepare/preprocess.py
code served (permissive licence) · get_code("9da7c93f348eb265")
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 parse_sent(line):
    delim = '[.|?|,|!|;]'
    sents = []
    segs = re.split(delim, line)
    for seg in segs:
        if seg not in delim and len(seg) > 5:
            sents.append(seg)
    return sents

ARGS = {
    'line': "Hello world. This is a test! What about this, right? Yes; indeed."
}
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
process_parsed_texts Ran
ran; our contract draft was wrong, not the code
BobXWu/ECRTM/ECRTM/preprocess/preprocess.py
pointer only (licence: NONE) · get_code("ed01133f9ff25fd3")
How it ran

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

purity Ran
ran; our contract draft was wrong, not the code
SmilesDZgk/GNTM/utils.py
code served (permissive licence) · get_code("e0797ad5fbc7e81b")
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 float64, ndim 0, 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 simple test data for the purity function
# labels: ground truth cluster assignments
# preds: predicted cluster assignments

# Create labels with 3 classes (0, 1, 2)
labels = np.array([0, 0, 1, 1, 2, 2])

# Create predictions with 2 clusters (0, 1)
# Mix them so there's some overlap but not perfect purity
preds = np.array([0, 0, 0, 1, 1, 1])

ARGS = {
    "labels": labels,
    "preds": preds
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
purity_score Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
BobXWu/ECRTM/ECRTM/utils/eva/cluster.py
pointer only (licence: NONE) · get_code("1645ca2fb55ce1f8")
How it ran

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

GraphConvolution Not yet run
unverified (harvested, no recorded run)
valdersoul/GraphBTM/layers.py
pointer only (licence: NONE) · get_code("f518c552a0f66557")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, valdersoul/GraphBTM 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. No invocation recorded.

todatapath Not yet run
unverified (harvested, no recorded run)
SmilesDZgk/GNTM/utils.py
code served (permissive licence) · get_code("7a666aa691ea4f90")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 11 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
purity_score 7 other papers
5 dated earlier, 2 later
2201.02609 (2022-01)
Generalized Category Discovery
2305.06144 (2023-05)
Learning Semi-supervised Gaussian Mixture Models for Generalized Category Discovery
and 4 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

Topic modeling is a widely used approach for analyzing and exploring large document collections. Recent research efforts have incorporated pre-trained contextualized language models, such as BERT embeddings, into topic modeling. However, they often neglect the intrinsic informational value conveyed by mutual dependencies between words. In this study, we introduce GINopic, a topic modeling framework based on graph isomorphism networks to capture the correlation between words. By conducting intrinsic (quantitative as well as qualitative) and extrinsic evaluations on diverse benchmark datasets, we demonstrate the effectiveness of GINopic compared to existing topic models and highlight its potential for advancing topic modeling.

For agents

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

get_harvested_code_for_paper("2404.02115")
get_code_for_paper("2404.02115")
have("2404.02115")

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

curl https://syntology.ai/api/ran/2404.02115.json

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

[![Syntology run record](https://syntology.ai/api/ran/2404.02115.svg)](https://syntology.ai/paper/2404.02115)

Connect an agent — have() is free.