MétaCan
Menu
← Back to cohort
Record W7132569310

Diversity is not a one-way street: pilot study on ethical interventions for racial bias in text-to-image systems

2023· article· en· W7132569310 on OpenAlexaffvenue
Kathleen Fraser, Sevtlana Kiritchenko, Isar Nejadgholi

Bibliographic record

VenueNPARC · 2023
Typearticle
Languageen
FieldSocial Sciences
TopicEthics and Social Impacts of AI
Canadian institutionsNational Research Council Canada
Fundersnot available
KeywordsDiversity (politics)Psychological interventionInclusion (mineral)PortraitWork (physics)Racial biasGender diversityRelevance (law)
DOInot available

Abstract

fetched live from OpenAlex

Text-to-image generation models can reflect the underlying societal biases present in their training data. However, user-level interventions to encourage greater diversity in the output have been proposed. Here, we examine visually stereotypical output from three widely-used models: DALL-E 2, Mid-journey, and Stable Diffusion. Some of the prompts we consider (e.g., “a photo portrait of a lawyer”) result in an underrepresentation of darker-skinned individuals in the output, while other prompts (e.g., “a photo portrait of a felon”) result in over-representation of darker-skinned individuals. We show that existing linguistic interventions serve to correct for under-representation to some degree, but in fact amplify the bias in cases of over-representation for all three systems. Further work is needed to develop effective methods to promote equity, diversity, and inclusion in the output of image generation systems.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.018
metaresearch head score (Gemma)0.080
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesScience and technology studies
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.998
Threshold uncertainty score0.098

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0180.080
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0010.000
Science and technology studies0.0020.002
Scholarly communication0.0020.003
Open science0.0010.002
Research integrity0.0010.002
Insufficient payload (model declined to judge)0.0070.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.393
GPT teacher head0.465
Teacher spread0.072 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

Study designObservational
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2023
Admission routes2
Has abstractyes

Explore more

Same venueNPARC→Same topicEthics and Social Impacts of AI→French-language works237,207→