SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2202.07603 · 2022

Fairness Indicators for Systematic Assessments of Visual Feature Extractors

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 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; 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/2202.07603.json · badge: /api/ran/2202.07603.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
facebookresearch/vissl extension (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
generate_dollar_street_map Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
facebookresearch/vissl/projects/fairness_indicators/geographical_diversity/inference_geographical_diversity.py
code served (permissive licence) · get_code("bd133604d32811bc")
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 torch
import numpy as np

torch.manual_seed(0)

# Create sample image paths
image_paths = [
    'data/chair/img001.jpg',
    'data/table/img002.jpg',
    'data/chair/img003.jpg',
]

# Create sample metadata
metadata = [
    {
        '_id': 'img001',
        'label': 'chair',
        'country': 'USA',
        'region': 'North America',
        'income': 50000,
        'url': 'http://example.com/img001.jpg'
    },
    {
        '_id': 'img002',
        'label': 'table',
        'country': 'India',
        'region': 'Asia',
        'income': 20000,
        'url': 'http://example.com/img002.jpg'
    },
    {
        '_id': 'img003',
        'label': 'chair',
        'country': 'Brazil',
        'region': 'South America',
        'income': 30000,
        'url': 'http://example.com/img003.jpg'
    },
]

# Create sample predictions (indices into id_to_label)
# Each prediction is a list of predicted class indices
predictions = [
    [0, 1, 2, 3, 4, 5, 6],  # predictions for img001
    [1, 0, 3, 2, 5, 4, 7],  # predictions for img002
    [0, 2, 1, 4, 3, 6, 8],  # predictions for img003
]

# Create id_to_label mapping
# Maps string indices to label names (can be lists or strings)
id_to_label = {
    '0': ['chair', 'seat'],
    '1': ['table', 'desk'],
    '2': ['lamp', 'light'],
    '3': ['bed', 'mattress'],
    '4': ['sofa', 'couch'],
    '5': ['cabinet', 'cupboard'],
    '6': ['shelf', 'rack'],
    '7': ['stool', 'bench'],
    '8': ['wardrobe', 'closet'],
}

# Create confidence scores
confidence_scores = [
    [0.95, 0.85, 0.75, 0.65, 0.55, 0.45, 0.35],  # scores for img001
    [0.92, 0.88, 0.78, 0.68, 0.58, 0.48, 0.38],  # scores for img002
    [0.90, 0.80, 0.70, 0.60, 0.50, 0.40, 0.30],  # scores for img003
]

topk = 5

ARGS = {
    'image_paths': image_paths,
    'metadata': metadata,
    'predictions': predictions,
    'id_to_label': id_to_label,
    'confidence_scores': confidence_scores,
    'topk': topk,
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
numpy, torch — 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
get_per_attribute_accuracy Ran
ran; our contract draft was wrong, not the code
facebookresearch/vissl/projects/fairness_indicators/geographical_diversity/inference_geographical_diversity.py
code served (permissive licence) · get_code("f4c07a269976ba8a")
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 torch
import numpy as np

torch.manual_seed(0)

# Build a filtered_input_map that matches the expected structure
# The function expects a dict where:
# - keys are identifiers (e.g., integers or strings)
# - values are dicts with keys: 'concept', 'prediction', 'income', 'url', 'raw_prediction', and potentially other attributes

filtered_input_map = {
    'item_0': {
        'concept': 'cat',
        'prediction': ['cat', 'dog'],
        'income': 50000,
        'url': 'http://example.com/0',
        'raw_prediction': [0.8, 0.2],
        'age': 25,
        'gender': 'male',
        'location': 'urban'
    },
    'item_1': {
        'concept': 'dog',
        'prediction': ['dog'],
        'income': 60000,
        'url': 'http://example.com/1',
        'raw_prediction': [0.1, 0.9],
        'age': 30,
        'gender': 'female',
        'location': 'rural'
    },
    'item_2': {
        'concept': 'bird',
        'prediction': ['cat', 'bird'],
        'income': 70000,
        'url': 'http://example.com/2',
        'raw_prediction': [0.5, 0.5],
        'age': 25,
        'gender': 'male',
        'location': 'urban'
    },
    'item_3': {
        'concept': 'cat',
        'prediction': ['dog', 'bird'],
        'income': 55000,
        'url': 'http://example.com/3',
        'raw_prediction': [0.3, 0.7],
        'age': 30,
        'gender': 'female',
        'location': 'suburban'
    }
}

ARGS = {
    'filtered_input_map': filtered_input_map,
    'predict_attribute': 'concept'
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
numpy, torch — 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 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 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 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.

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

Does everyone equally benefit from computer vision systems? Answers to this question become more and more important as computer vision systems are deployed at large scale, and can spark major concerns when they exhibit vast performance discrepancies between people from various demographic and social backgrounds. Systematic diagnosis of fairness, harms, and biases of computer vision systems is an important step towards building socially responsible systems. To initiate an effort towards standardized fairness audits, we propose three fairness indicators, which aim at quantifying harms and biases of visual systems. Our indicators use existing publicly available datasets collected for fairness evaluations, and focus on three main types of harms and bias identified in the literature, namely harmful label associations, disparity in learned representations of social and demographic traits, and biased performance on geographically diverse images from across the world.We define precise experimental protocols applicable to a wide range of computer vision models. These indicators are part of an ever-evolving suite of fairness probes and are not intended to be a substitute for a thorough analysis of the broader impact of the new computer vision technologies. Yet, we believe it is a necessary first step towards (1) facilitating the widespread adoption and mandate of the fairness assessments in computer vision research, and (2) tracking progress towards building socially responsible models. To study the practical effectiveness and broad applicability of our proposed indicators to any visual system, we apply them to off-the-shelf models built using widely adopted model training paradigms which vary in their ability to whether they can predict labels on a given image or only produce the embeddings. We also systematically study the effect of data domain and model size.

For agents

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

get_harvested_code_for_paper("2202.07603")
get_code_for_paper("2202.07603")
have("2202.07603")

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

curl https://syntology.ai/api/ran/2202.07603.json

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

[![Syntology run record](https://syntology.ai/api/ran/2202.07603.svg)](https://syntology.ai/paper/2202.07603)

Connect an agent — have() is free.