Pith. sign in

REVIEW 5 major objections 6 minor 27 references

INFELM: In-depth Fairness Evaluation of Large Text-To-Image Models

T0 review · 5 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read INFELM claims a facial-topology skintone classifier exposes that nine text-to-image models fail a common fairness rule when bias is measured accurately.

desk verdict The skintone classifier is a genuine step forward, but every headline fairness number in Table 7 depends on that classifier's unmeasured accuracy on the very models being audited. read the letter →

arxiv 2501.01973 v3 pith:VOPBWKXS submitted 2024-12-28 cs.CV cs.AIcs.CY

Di Jin , Xing Liu , Yu Liu , Jia Qing Yap , Andrea Wong , Adriana Crespo , Qi Lin , Zhiyuan Yin
show 2 more authors
Qiang Yan Ryan Ye
This is my paper · ORCID
classification cs.CVcs.AIcs.CY
keywords fairnessevaluationtext-to-imagemodelsskintoneclassificationrepresentationbiascontentalignmentfour-fifthrulefacialtopology
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that existing text-to-image fairness audits are unreliable because their skintone classifiers rely on detecting dominant pixel colors and mapping to the nearest Monk scale, which misclassifies images due to uneven color spaces and lighting. It proposes INFELM, a fairness evaluation framework with a new skintone classifier that combines facial topological features with a distribution of dominant skin pixels, and reports that this classifier beats pixel-only baselines by at least 16.04% in precision. Using this measurement on nine text-to-image models across six bias-sensitive domains, the paper finds that most models fail an empirical four-fifth rule fairness criterion, with representation bias more pronounced than content alignment errors and skintone alignment errors substantially higher than gender alignment errors. The upshot is a reusable benchmark and a warning that current models are not meeting basic demographic fairness standards when measured more accurately.

What carries the argument

The central mechanism is the INFELM skintone classifier. It first trains a small CNN on 12,000 synthetically generated faces (using RealisticVision v5.1 across six geographic-origin groups, with varying environment lighting) to extract latent facial topological features, then extracts the K = 15 most dominant skin pixels in the facial region via Otsu's thresholding in YCbCr and HSV color spaces, representing them as a bin-weighted distribution over Monk scales. These two feature types are fused through a self-attention module and trained with a loss L = alpha*L_ft + (1-alpha)*L_st as an ordinal classification problem with ±1 scale tolerance. Fairness is then scored with representation bias b = (1/Z)*sum_g |n_g/N - p_g| normalized to the maximum possible bias, and content alignment error as the prompt-classifier mismatch ratio, both evaluated against a uniform reference distribution and a four-fifth rule threshold of 0.2.

What would settle it

Take the same 246 prompts and generated images, but compute representation bias using real-world social statistics for each domain (e.g., occupation gender shares from labor statistics, skintone proportions from census-like data) instead of uniform p_g; if the four-fifth rule failures disappear or reorder, the audit's conclusion depends on that uniform assumption rather than on the models' outputs.

Watch

Extended reading notes

Core claim

The central discovery is that a skintone classifier which fuses learned facial topological features with an ordinal distribution of dominant skin pixels (rather than a single mean color) reduces skintone misclassification, and that this measurement change affects the assessment of large text-to-image models: none of the nine models evaluated consistently satisfies the four-fifth rule threshold b < 0.2 for representation bias. Across all models, average gender representation bias is 0.639 and skintone bias 0.593, while average gender alignment error is only 0.029 versus 0.668 for skintone, so the dominant failure is inaccurate skintone alignment rather than gender misalignment. The paper also finds that fine-tuned photorealistic models (RealisticVision variants) show more polarized demographics than their base Stable Diffusion models, and that DALL-E 3 is closest to the fairness criteria but still fails on skintone.

Load-bearing premise

All bias scores assume demographic groups should be equally represented (p_g uniform); if the true reference population is non-uniform, every bias number changes.

Editorial extensions

If this is right

  • If the classifier is adopted, downstream fairness audits become more trustworthy because skintone measurements are no longer skewed by uneven Monk-scale color spans or lighting.
  • Most evaluated models fail the four-fifth rule; practitioners should not assume a single model is fair across both gender and skintone.
  • Fine-tuned photorealistic models tend to show more polarized demographic outputs, suggesting fine-tuning on curated datasets can amplify bias.
  • Skintone alignment errors exceed gender errors by a large margin, pointing to missing or noisy skintone captions in training data as a lever for improvement.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The uniform reference distribution is a policy choice, not a fact; the same audit pipeline would yield different bias scores under a reference distribution drawn from population statistics, so the numerical thresholds should be read as relative rather than absolute.
  • The claim that fine-tuning amplifies bias is based on only two RealisticVision variants; a broader sample of fine-tuned models would be needed to confirm the trend.
  • The synthetic training images for the topology classifier come from a single text-to-image model (RealisticVision v5.1); if that generator has its own style bias, the latent topological features may inherit it, which is testable by retraining on a different generator.
  • The content alignment error metric counts any mismatch between prompt and classifier as an error, but some mismatches may be legitimate (e.g., a prompt with no demographic attribute); refining the metric could change the error magnitudes.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. The paper presents INFELM, a fairness evaluation framework for large text-to-image (T2I) models. It proposes a skintone classifier that fuses latent facial-topology features with a dominant-skin-pixel distribution, and evaluates content alignment error and representation bias across gender and five skintone groups. Using 246 prompts in six socially sensitive domains and 100 generated images per prompt, the authors compare nine models (Stable Diffusion v1.4/1.5/2.1, Openjourney v4, DALL-E 3, RealisticVision v5.1/6.0, SDXL Lightning, FLUX.1-schnell). They report that the classifier improves skintone precision by at least 16.04 percentage points over HEIM and VIT baselines, and that most evaluated models fail the four-fifth rule, with representation bias generally larger than alignment error and skintone alignment error substantially larger than gender error.

Significance. If the measurement assumptions hold, the paper would provide a useful reusable fairness benchmark and a convincing demonstration that pixel-only skintone detection is insufficient. The large-scale comparison across nine models and six domains, the explicit treatment of the Monk scale's uneven color distribution, and the inclusion of both representation and alignment metrics are genuine strengths. The classifier results on the public WBB dataset are encouraging. However, the headline fairness conclusions rest on an unvalidated generalization of the skintone classifier to the evaluated models, an untested uniform reference distribution, and an underspecified alignment-error mapping; these issues must be resolved before the empirical claims can be taken as established.

major comments (5)
  1. [Section 5.2, Table 7] All skintone metrics for the nine evaluated models are computed using the INFELM classifier, but the classifier is never validated on images produced by these models. The only evaluations are on WBB and High-Aes (Table 6), and the topology module is trained on synthetic faces generated by RealisticVision v5.1 (Section 4.2.1), which is itself one of the nine evaluated models. Since style shifts or group-correlated misclassification (e.g., adjacent Monk-scale confusion under stylized lighting) would change every value of b_s and e_s in Table 7, the headline conclusions in the Abstract and Takeaway 1 are not supported without per-model human-annotated validation or an explicit style-robustness analysis.
  2. [Section 4.3.2, Eq. (4)] The representation bias metric defines the reference distribution by the statement 'we assume that the demographic groups are equally distributed.' No justification or sensitivity analysis is provided, even though the text immediately acknowledges that social statistics from authoritative sources could serve as the groundtruth. Because the four-fifth-rule checks in Section 5.3 compare every model to this uniform baseline, all bias scores in Table 7 and the conclusion that 'most models do not meet the criteria of fairness' are conditional on an untested normative choice. The authors should either test alternative reference distributions or explicitly reframe the results as an equal-representation benchmark rather than a fairness verdict.
  3. [Section 4.3.2, alignment error definition] The mismatch ratio used to compute e_g and e_s is not defined formally, and no mapping is given from the demographic phrases in the prompts (e.g., 'a black CEO') to the 10 Monk-scale labels produced by the classifier. Without this mapping, the large skintone alignment errors in Table 7 are uninterpretable; the claim that skintone alignment error is 'significantly higher' than gender error (Takeaway 2) could be an artifact of an inconsistent or overly strict mapping rather than a property of the generators. The paper should specify the mapping, give examples, and ideally validate the alignment labels by human raters.
  4. [Section 5.1, Table 6] The High-Aes dataset is described only as an internal synthetic dataset of 11,000 images with human annotations. It is not stated whether these images are generated with the same RealisticVision v5.1 pipeline used to create the topology training data in Section 4.2.1, nor are class balance, annotation protocol, or the test split described. Without this information, the reported 0.9033 precision on High-Aes and the claimed 'at least 16.04%' improvement over HEIM cannot be independently assessed, and the comparison may not be a fair out-of-distribution test.
  5. [Section 5.2, Table 7] All model comparisons are point estimates with no confidence intervals, standard errors, or significance tests, despite the finite sample sizes (100 images per prompt, 246 prompts). For example, differences as small as 0.001 in e_g between SDXL Lightning and FLUX.1-schnell are treated as meaningful without uncertainty quantification. This undermines the comparative claims in Takeaway 4; bootstrap confidence intervals or per-prompt standard errors should be reported.
minor comments (6)
  1. [Section 2] The typos 'evaluaiton' and 'exsting' should be corrected to 'evaluation' and 'existing'; the Related Work section should be copy-edited throughout.
  2. [Section 4.2.3] 'Ordinary classification problem' should read 'ordinal classification problem,' and the sentence explaining the relationship between tolerance and ordinal classification should be rewritten for clarity.
  3. [Section 5.3] The sentence 'compare them with the fairness reference computed using the empirical four-fifth rule, .' contains an empty clause and appears to be missing a phrase or cross-reference to Figure 6.
  4. [Equation (3)] The symbol alpha is described as 'the attention weights' but the equation uses it as a scalar balancing coefficient; clarify whether alpha is a learned attention vector or a fixed scalar loss weight.
  5. [Section 5.2] 'Off alltext-to-image models' should read 'Of all text-to-image models'; the text also alternates between 'DALL-E 3' and 'Dall-E 3' and should be made consistent.
  6. [Reproducibility] The paper does not provide a link to code, the full prompt set, or generated-image samples; releasing these artifacts would substantially improve reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the fairness evaluation is self-contained against external benchmarks, with only a tooling-level validity caveat.

full rationale

The paper's claimed derivation chain is: construct bias-sensitive prompts, generate images with nine text-to-image models, label the images with a gender classifier and the INFELM skintone classifier, then compute representation bias and content alignment error against a stated reference distribution. No step makes the predicted fairness outcome equivalent to an input by construction. The skintone classifier is trained partly on synthetic faces generated by RealisticVision v5.1 (Section 4.2.1), which is one of the nine evaluated models, and the paper does not report per-model accuracy of the classifier on the other models' outputs. This is a genuine external-validity limitation: a style-shift or group-correlated classifier error could shift the observed skintone bias and alignment values in Table 7. However, it is not a circular reduction, because the classifier is independently benchmarked on the external WBB dataset and on a held-out High-Aes test set (Table 6), and the fairness scores are not fitted parameters derived from the models' outputs. The uniform demographic reference distribution (Section 4.3.2) is an explicit normative assumption rather than a self-referential definition, and the four-fifth-rule thresholds are external criteria. There are no load-bearing self-citations and no imported uniqueness theorems. The central findings are therefore not forced by the paper's inputs; the open question is measurement validity under style shift, not circularity.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The representation-bias and alignment conclusions depend on the uniform demographic baseline, the 0.2 threshold, and the in-house classifier; these are the main things the reader pays for that are not derived or externally grounded.

free parameters (4)
  • fairness threshold epsilon = 0.2
    Selected to operationalize the four-fifth rule in Eq. (2) and Section 5.3; used to declare a model biased or misaligned. No sensitivity analysis.
  • dominant pixel count K = 15
    Number of dominant skin pixels retained for the distribution representation in Section 4.2.2; no ablation reported.
  • Monk scale error tolerance = ±1 scale
    Used to define ordinal classification acceptability in Section 4.2.3; affects training loss and evaluation of the classifier, but not the central fairness numbers.
  • synthetic training data size per group = 2000 train, 400 test per group across 6 groups
    Dataset size for topology classifier in Section 4.2.1; no sensitivity analysis or cross-validation.
assumptions (4)
  • domain assumption Demographic groups are equally distributed in the target use cases.
    Section 4.3.2 states this explicitly; it is normative and drives all representation-bias results. The paper mentions social statistics but does not use them.
  • domain assumption The four-fifth rule can be represented as b < 0.2 and e < 0.2.
    Section 5.3 applies the threshold 0.2 to normalized total-variation bias and to alignment error; this is an extension of the legal disparate-impact rule, not a derivation from it.
  • domain assumption Classifier predictions can be treated as ground-truth demographic labels for generated images.
    Sections 4.3 and 5.2 compute alignment errors and bias from gender and skintone classifiers, with no human validation on the generated test images.
  • domain assumption Synthetic faces generated by RealisticVision v5.1 are representative of the six OMB-derived groups for learning facial topology.
    Section 4.2.1 uses RealisticVision-generated images to train the topology classifier; if these images do not cover real facial diversity, the classifier will inherit the generator's biases.

how reviews work

0 comments
Cite this review

Pith. "Pith review of INFELM: In-depth Fairness Evaluation of Large Text-To-Image Models." pith.science (2026). https://pith.science/paper/VOPBWKXS

@misc{pith2026250101973,
  author       = {Pith},
  title        = {Pith review of: INFELM: In-depth Fairness Evaluation of Large Text-To-Image Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VOPBWKXS}},
  note         = {Machine review of arXiv:2501.01973}
}
read the original abstract

The rapid development of large language models (LLMs) and large vision models (LVMs) have propelled the evolution of multi-modal AI systems, which have demonstrated the remarkable potential for industrial applications by emulating human-like cognition. However, they also pose significant ethical challenges, including amplifying harmful content and reinforcing societal biases. For instance, biases in some industrial image generation models highlighted the urgent need for robust fairness assessments. Most existing evaluation frameworks focus on the comprehensiveness of various aspects of the models, but they exhibit critical limitations, including insufficient attention to content generation alignment and social bias-sensitive domains. More importantly, their reliance on pixel-detection techniques is prone to inaccuracies. To address these issues, this paper presents INFELM, an in-depth fairness evaluation on widely-used text-to-image models. Our key contributions are: (1) an advanced skintone classifier incorporating facial topology and refined skin pixel representation to enhance classification precision by at least 16.04%, (2) a bias-sensitive content alignment measurement for understanding societal impacts, (3) a generalizable representation bias evaluation for diverse demographic groups, and (4) extensive experiments analyzing large-scale text-to-image model outputs across six social-bias-sensitive domains. We find that existing models in the study generally do not meet the empirical fairness criteria, and representation bias is generally more pronounced than alignment errors. INFELM establishes a robust benchmark for fairness assessment, supporting the development of multi-modal AI systems that align with ethical and human-centric principles.

Figures

Figures reproduced from arXiv: 2501.01973 by the authors.

Figure 1
Figure 1. Skintone confusion. Due to the unevenly dis￾tributed color spectrum and lighting disturbance, color￾extraction and shortest-distance-based methods incorrectly group facial images into similar groups, which impedes the correctness of downstream analysis. In this example, most images are incorrectly labeled with Monk scale 6 due to its dominance in the Euclidean-represented RGB space of skin￾tones. in total 6 target d… view at source ↗
Figure 2
Figure 2. INFELM overview. Given a bias-sensitive domain with fairness risks (e.g. wealth, education), INFELM first feeds the prompts following the groundtruth distribution to a text-to-image model to generate images in scale. Then, the demographics classifiers automatically generate the corresponding labels, which are used for fairness analysis. In the end, INFELM outputs the comprehensive fairness analysis and the bias scor… view at source ↗
Figure 3
Figure 3. Highlight of skintone pixel representations: mean, dominant pixel distributions before Otsu’s method. The ag￾gregated mean pixel representation makes it less distinguish￾able while the ordinal distributions (ordered by pixel weights) preserve more detailed information. 4.2.3 Skintone classification. INFELM takes the outputs from the above two components, and then performs fusion on the aforementioned features as the… view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Skintone classification model architecture based on multi-modality feature fusion classification and aligns with the ordinal nature of the MST Scale (from the lightest to darkest), where adjacent categories are closely related. The ordinary classification problem can a…
Figure 5
Figure 5. Figure 5: Fairness analysis of Stable Diffusion v1.4 skintone information during training or fine-tuning affects the generation alignment. This also indicates that we may see similar issues in other models trained or fine-tuned with the same set of data corpus. Overall, we find …
Figure 6
Figure 6. Figure 6: Fairness degree of industry benchmark models applications, practitioners should be aware of these dispari￾ties before adopting a specific model to mitigate the potential negative impacts of generated content. 6 Conclusion Governing fairness in text-to-image models is n…
Figure 8
Figure 8. Figure 8: Complete fairness analysis per text-to-image model [PITH_FULL_IMAGE:figures/full_fig_p010_8.png]
Figure 8
Figure 8. Figure 8: (Continued) Complete fairness analysis per text-to-image model [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 18 canonical work pages

  1. [1]

    https://huggingface.co/touchtech/fashion-images- gender-age-vit-large-patch16-224-in21k-v3

    gender-age-vit. https://huggingface.co/touchtech/fashion-images- gender-age-vit-large-patch16-224-in21k-v3 . Accessed: 2024-10-2

    work page 2024

  2. [2]

    https://huggingface.co/prompthero/openjourney-v4

    Openjourney v4. https://huggingface.co/prompthero/openjourney-v4. Accessed: 2024-10-2

    work page 2024

  3. [3]

    https://huggingface.co/SG161222/Realistic_ Vision_V5.1_noVAE

    Realisticvision v5.1. https://huggingface.co/SG161222/Realistic_ Vision_V5.1_noVAE. Accessed: 2024-10-2

    work page 2024

  4. [4]

    https://huggingface.co/SG161222/Realistic_ Vision_V6.0_B1_noVAE

    Realisticvision v6.0. https://huggingface.co/SG161222/Realistic_ Vision_V6.0_B1_noVAE. Accessed: 2024-10-2

    work page 2024

  5. [5]

    Ai fairness 360: An extensible toolkit for detecting and mitigating algorithmic bias

    Rachel KE Bellamy, Kuntal Dey, Michael Hind, Samuel C Hoffman, Stephanie Houde, Kalapriya Kannan, Pranay Lohia, Jacquelyn Martino, Sameep Mehta, Aleksandra Mojsilović, et al. Ai fairness 360: An extensible toolkit for detecting and mitigating algorithmic bias. IBM Journal of Research and Development , 63(4/5):4–1, 2019

    work page 2019

  6. [6]

    Easily accessible text-to-image generation amplifies demographic stereotypes at large scale

    Federico Bianchi, Pratyusha Kalluri, Esin Durmus, Faisal Ladhak, Myra Cheng, Debora Nozza, Tatsunori Hashimoto, Dan Jurafsky, James Zou, and Aylin Caliskan. Easily accessible text-to-image generation amplifies demographic stereotypes at large scale. In Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency , pages 1493–1504, 2023

    work page 2023

  7. [7]

    Dall-eval: Probing the reasoning skills and social biases of text-to-image generation models

    Jaemin Cho, Abhay Zala, and Mohit Bansal. Dall-eval: Probing the reasoning skills and social biases of text-to-image generation models. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 3043–3054, 2023

    work page 2023

  8. [8]

    Bias and fairness in large language models: A survey

    Isabel O Gallegos, Ryan A Rossi, Joe Barrow, Md Mehrab Tanjim, Sungchul Kim, Franck Dernoncourt, Tong Yu, Ruiyi Zhang, and Nes- reen K Ahmed. Bias and fairness in large language models: A survey. Computational Linguistics, pages 1–79, 2024

    2024

Show all 27 references
  1. [9]

    Gen- erative adversarial networks

    Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Gen- erative adversarial networks. Communications of the ACM, 63(11):139– 144, 2020

    2020

  2. [10]

    The pursuit of fairness in artificial intelligence models: A survey.arXiv preprint arXiv:2403.17333, 2024

    Tahsin Alamgir Kheya, Mohamed Reda Bouadjenek, and Sunil Aryal. The pursuit of fairness in artificial intelligence models: A survey.arXiv preprint arXiv:2403.17333, 2024

    2024 arXiv

  3. [11]

    Flux.1 [schnell]

    Black Forest Labs. Flux.1 [schnell]. https://huggingface.co/black- forest-labs/FLUX.1-schnell, 2024

    2024

  4. [12]

    Holistic evaluation of text-to-image models

    Tony Lee, Michihiro Yasunaga, Chenlin Meng, Yifan Mai, Joon Sung Park, Agrim Gupta, Yunzhi Zhang, Deepak Narayanan, Hannah Teufel, Marco Bellagente, et al. Holistic evaluation of text-to-image models. Advances in Neural Information Processing Systems , 36, 2024

    2024

  5. [13]

    Holistic evaluation of language models

    Percy Liang, Rishi Bommasani, Tony Lee, Dimitris Tsipras, Dilara Soylu, Michihiro Yasunaga, Yian Zhang, Deepak Narayanan, Yuhuai Wu, Ananya Kumar, et al. Holistic evaluation of language models. arXiv preprint arXiv:2211.09110, 2022

    2022 arXiv

  6. [14]

    Sdxl-lightning: Progres- sive adversarial diffusion distillation

    Shanchuan Lin, Anran Wang, and Xiao Yang. Sdxl-lightning: Progres- sive adversarial diffusion distillation. arXiv preprint arXiv:2402.13929, 2024

    2024 arXiv

  7. [15]

    Trustworthy llms: A survey and guideline for evaluating large language models’ alignment

    Yang Liu, Yuanshun Yao, Jean-Francois Ton, Xiaoying Zhang, Ruocheng Guo, Hao Cheng, Yegor Klochkov, Muhammad Faaiz Taufiq, and Hang Li. Trustworthy llms: A survey and guideline for evaluating large language models’ alignment. arXiv preprint arXiv:2308.05374 , 2023

    2023 arXiv

  8. [16]

    A survey on bias and fairness in machine learning

    Ninareh Mehrabi, Fred Morstatter, Nripsuta Saxena, Kristina Lerman, and Aram Galstyan. A survey on bias and fairness in machine learning. ACM computing surveys (CSUR), 54(6):1–35, 2021

    2021

  9. [17]

    The monk skin tone scale

    Ellis Monk. The monk skin tone scale. 2019

    2019

  10. [18]

    Dall-e 3: The latest in text-to-image generation

    OpenAI. Dall-e 3: The latest in text-to-image generation. https: //openai.com/blog/dall-e-3 , 2023. Accessed: 2024-10-2

    2023

  11. [19]

    Disentan- gling and operationalizing ai fairness at linkedin

    Joaquin Quiñonero Candela, Yuwen Wu, Brian Hsu, Sakshi Jain, Jen- nifer Ramos, Jon Adams, Robert Hallman, and Kinjal Basu. Disentan- gling and operationalizing ai fairness at linkedin. In Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency , pa...

    2023

  12. [20]

    Learning transferable visual models from natural language supervision

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learning transferable visual models from natural language supervision. In International conference on machine learning, pa...

    2021

  13. [21]

    Zero-shot text-to-image generation

    Aditya Ramesh, Mikhail Pavlov, Gabriel Goh, Scott Gray, Chelsea Voss, Alec Radford, Mark Chen, and Ilya Sutskever. Zero-shot text-to-image generation. In International conference on machine learning , pages 8821–8831. Pmlr, 2021

    2021

  14. [22]

    High-resolution image synthesis with latent diffu- sion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. High-resolution image synthesis with latent diffu- sion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 10684–10695, 2022

    2022

  15. [23]

    Photorealistic text-to-image diffu- sion models with deep language understanding

    Chitwan Saharia, William Chan, Saurabh Saxena, Lala Li, Jay Whang, Emily L Denton, Kamyar Ghasemipour, Raphael Gontijo Lopes, Burcu Karagol Ayan, Tim Salimans, et al. Photorealistic text-to-image diffu- sion models with deep language understanding. Advances in neural informati...

    2022

  16. [24]

    Aequitas: A bias and fairness audit toolkit

    Pedro Saleiro, Benedict Kuester, Loren Hinkson, Jesse London, Abby Stevens, Ari Anisfeld, Kit T Rodolfa, and Rayid Ghani. Aequitas: A bias and fairness audit toolkit. arXiv preprint arXiv:1811.05577, 2018

    2018 arXiv

  17. [25]

    Attngan: Fine-grained text to im- age generation with attentional generative adversarial networks

    Tao Xu, Pengchuan Zhang, Qiuyuan Huang, Han Zhang, Zhe Gan, Xiaolei Huang, and Xiaodong He. Attngan: Fine-grained text to im- age generation with attentional generative adversarial networks. In Proceedings of the IEEE conference on computer vision and pattern recog- nition, pa...

    2018

  18. [26]

    Scaling autoregres- sive models for content-rich text-to-image generation

    Jiahui Yu, Yuanzhong Xu, Jing Yu Koh, Thang Luong, Gunjan Baid, Zirui Wang, Vijay Vasudevan, Alexander Ku, et al. Scaling autoregres- sive models for content-rich text-to-image generation. arXiv preprint arXiv:2206.10789, 2(3):5, 2022

    2022 arXiv

  19. [27]

    Stackgan: Text to photo- realistic image synthesis with stacked generative adversarial networks

    Han Zhang, Tao Xu, Hongsheng Li, Shaoting Zhang, Xiaogang Wang, Xiaolei Huang, and Dimitris N Metaxas. Stackgan: Text to photo- realistic image synthesis with stacked generative adversarial networks. In Proceedings of the IEEE international conference on computer vision , page...

    2017

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.