SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.07183 · 2022

Visual Classification via Description from Large Language Models

arXiv · PDF · Open in the Atlas

Code that ran

6 of 6 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, 0 violated it and 0 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/2210.07183.json · badge: /api/ran/2210.07183.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
sachit-menon/classify_by_description_release canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
explainableml/waffleclip extension (classifier label)
classifier basis: classifier_2025
3 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
aggregate_similarity Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
sachit-menon/classify_by_description_release/load.py
pointer only (licence: NONE) · get_code("3936ee935ec8fce1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, sachit-menon/classify_by_description_release 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.

denormalize Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
explainableml/waffleclip/waffle_tools.py
code served (permissive licence) · get_code("7496518009557dce")
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 Tensor, ndim 4, dtype family float, shape (2, 3, 4, 5) 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 batch of normalized images
# Shape: (batch_size=2, channels=3, height=4, width=5)
images = torch.randn(2, 3, 4, 5)

# Use default means and stds (ImageNet normalization values)
means = (0.485, 0.456, 0.406)
stds = (0.229, 0.224, 0.225)

ARGS = {
    'images': images,
    'means': means,
    'stds': stds
}
Setup block drafted by
a language model, Claude Sonnet 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, 3, 4, 5) — shape (2, 3, 4, 5) 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
generate_prompt Ran
ran; our contract draft was wrong, not the code
explainableml/waffleclip/generate_descriptors.py
code served (permissive licence) · get_code("d382e047433f0947")
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 str. 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.

ARGS = {
    "category_name": "dog"
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
generate_prompt Ran
ran; our contract draft was wrong, not the code
sachit-menon/classify_by_description_release/generate_descriptors.py
pointer only (licence: NONE) · get_code("ea60dc7cfafc02ad")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, sachit-menon/classify_by_description_release 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.

load_json Ran
ran; our contract draft was wrong, not the code
explainableml/waffleclip/waffle_tools.py
code served (permissive licence) · get_code("3de9cd37f5734894")
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 json
import os

# Create a temporary JSON file with some content
temp_dir = tempfile.mkdtemp()
json_filename = os.path.join(temp_dir, "test_file.json")

# Write some JSON data to the file
test_data = {"key": "value", "number": 42, "list": [1, 2, 3]}
with open(json_filename, 'w') as f:
    json.dump(test_data, f)

ARGS = {
    "filename": json_filename
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
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 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
stringtolist Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from tmlr-group/eoe
pointer only; the stored copy get_code serves, from tmlr-group/eoe, is pointer only (licence: NONE) · get_code("46062947cf3ab0e6")
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.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 5 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 5 later) and 1 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
stringtolist 6 other papers
6 later
2306.07282 (2023-06)
Waffling around for Performance: Visual Classification with Random Words and Broad Concepts
2404.00419 (2024-04)
Do Vision-Language Models Understand Compound Nouns?
2406.00806 (2024-06)
Envisioning Outlier Exposure by Large Language Models for Out-of-Distribution Detection
and 3 more not named here
aggregate_similarity 2 other papers
2 later
2407.14412 (2024-07)
DEAL: Disentangle and Localize Concept-level Explanations for VLMs
2504.20860 (2025-04)
FedMVP: Federated Multimodal Visual Prompt Tuning for Vision-Language Models
load_json 2 other papers
2 later
2306.07282 (2023-06)
Waffling around for Performance: Visual Classification with Random Words and Broad Concepts
denormalize 1 other paper
1 later
2306.07282 (2023-06)
Waffling around for Performance: Visual Classification with Random Words and Broad Concepts
generate_prompt 1 other paper
1 later
2306.07282 (2023-06)
Waffling around for Performance: Visual Classification with Random Words and Broad Concepts

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

Vision-language models (VLMs) such as CLIP have shown promising performance on a variety of recognition tasks using the standard zero-shot classification procedure -- computing similarity between the query image and the embedded words for each category. By only using the category name, they neglect to make use of the rich context of additional information that language affords. The procedure gives no intermediate understanding of why a category is chosen, and furthermore provides no mechanism for adjusting the criteria used towards this decision. We present an alternative framework for classification with VLMs, which we call classification by description. We ask VLMs to check for descriptive features rather than broad categories: to find a tiger, look for its stripes; its claws; and more. By basing decisions on these descriptors, we can provide additional cues that encourage using the features we want to be used. In the process, we can get a clear idea of what features the model uses to construct its decision; it gains some level of inherent explainability. We query large language models (e.g., GPT-3) for these descriptors to obtain them in a scalable way. Extensive experiments show our framework has numerous advantages past interpretability. We show improvements in accuracy on ImageNet across distribution shifts; demonstrate the ability to adapt VLMs to recognize concepts unseen during training; and illustrate how descriptors can be edited to effectively mitigate bias compared to the baseline.

For agents

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

get_harvested_code_for_paper("2210.07183")
get_code_for_paper("2210.07183")
have("2210.07183")

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

curl https://syntology.ai/api/ran/2210.07183.json

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

[![Syntology run record](https://syntology.ai/api/ran/2210.07183.svg)](https://syntology.ai/paper/2210.07183)

Connect an agent — have() is free.