SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2107.02173 · NeurIPS · 2021

Is Automated Topic Model Evaluation Broken?: The Incoherence of Coherence

Jordan Boyd-Graber, Alexander Hoyle, Pranav Goel, Denis Peskov, Andrew Hian-Cheong, Philip Resnik

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 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, 0 violated it and 0 have no contract verdict; 2 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/2107.02173.json · badge: /api/ran/2107.02173.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
ahoho/topics canonical (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
FunctionStatusWhere it lives
load_estimates Ran
ran; our contract draft was wrong, not the code
ahoho/topics/runs/calculate_metrics.py
code served (permissive licence) · get_code("fa434595d25180ad")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) 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 tempfile
import shutil
from pathlib import Path
import numpy as np

# Create a temporary directory structure for the model
temp_dir = tempfile.mkdtemp()
model_dir = Path(temp_dir) / "experiments" / "dvae" / "model_001"
model_dir.mkdir(parents=True, exist_ok=True)

# Create dummy data files for dvae model type
# topic_word: (num_topics, vocab_size)
num_topics = 3
vocab_size = 5
num_train_docs = 4
num_test_docs = 2

# For dvae, beta is saved as logits (will be softmaxed)
beta = np.random.randn(num_topics, vocab_size).astype(np.float32)
np.save(model_dir / 'beta.npy', beta)

# doc_topic distributions: (num_docs, num_topics)
train_theta = np.random.rand(num_train_docs, num_topics).astype(np.float32)
train_theta = train_theta / train_theta.sum(axis=1, keepdims=True)
np.save(model_dir / 'train.theta.npy', train_theta)

test_theta = np.random.rand(num_test_docs, num_topics).astype(np.float32)
test_theta = test_theta / test_theta.sum(axis=1, keepdims=True)
np.save(model_dir / 'test.theta.npy', test_theta)

ARGS = {
    'model_dir': str(model_dir),
    'model_type': None  # Will be inferred from path
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
numpy, scipy — 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
load_mallet_doc_topic Ran
ran, honoured the contract we drafted
ahoho/topics/runs/calculate_metrics.py
code served (permissive licence) · get_code("de0767b68b91b262")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned ndarray, ndim 2, dtype family float, shape (2, 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 tempfile
import os

# Create a temporary file with mallet doc-topic format
# Format: doc_id doc_name topic1_prob topic2_prob topic3_prob ...
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')

# Write sample mallet doc-topic data
# First line might be a header (if skip_first=True)
temp_file.write("#doc source topic proportions...\n")
temp_file.write("0\tdoc0.txt\t0.5\t0.3\t0.2\n")
temp_file.write("1\tdoc1.txt\t0.1\t0.6\t0.3\n")
temp_file.close()

ARGS = {
    'fpath': temp_file.name,
    'skip_first': True
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
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, 3) — shape (2, 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
load_yaml Ran
ran; our contract draft was wrong, not the code
ahoho/topics/runs/calculate_metrics.py
code served (permissive licence) · get_code("08bfbf35112d40e0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) 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 tempfile
import os
import yaml

# Create a temporary YAML file
temp_dir = tempfile.mkdtemp()
yaml_path = os.path.join(temp_dir, "test_config.yaml")

# Write some sample YAML content
yaml_content = {
    'model': 'test_model',
    'batch_size': 32,
    'learning_rate': 0.001,
    'layers': [64, 128, 256],
    'nested': {
        'key1': 'value1',
        'key2': 42
    }
}

with open(yaml_path, 'w') as f:
    yaml.dump(yaml_content, f)

ARGS = {
    'path': yaml_path
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
yaml — 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

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 3 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.

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 model evaluation, like evaluation of other unsupervised methods, can be contentious. However, the field has coalesced around automated estimates of topic coherence, which rely on the frequency of word co-occurrences in a reference corpus. Contemporary neural topic models surpass classical ones according to these metrics. At the same time, topic model evaluation suffers from a validation gap: automated coherence, developed for classical models, has not been validated using human experimentation for neural models. In addition, a meta-analysis of topic modeling literature reveals a substantial standardization gap in automated topic modeling benchmarks. To address the validation gap, we compare automated coherence with the two most widely accepted human judgment tasks: topic rating and word intrusion. To address the standardization gap, we systematically evaluate a dominant classical model and two state-of-the-art neural models on two commonly used datasets. Automated evaluations declare a winning model when corresponding human evaluations do not, calling into question the validity of fully automatic evaluations independent of human judgments.

For agents

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

get_harvested_code_for_paper("2107.02173")
get_code_for_paper("2107.02173")
have("2107.02173")

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

curl https://syntology.ai/api/ran/2107.02173.json

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

[![Syntology run record](https://syntology.ai/api/ran/2107.02173.svg)](https://syntology.ai/paper/2107.02173)

Connect an agent — have() is free.