Pith. sign in

REVIEW 4 major objections 7 minor 68 references

EMoE: Training-Free Expert Disagreement for Uncertainty-Aware Text-to-Image Diffusion

T0 review · 4 major / 7 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read A training-free variance signal inside diffusion MoEs predicts when a prompt will generate poorly.

desk verdict Training-free MoE expert-variance uncertainty is a genuinely new and practical tool, and the empirical trend is real, but the epistemic-uncertainty interpretation is overclaimed and a prompt-length confound remains untested. read the letter →

arxiv 2505.13273 v2 pith:KYFG42MW submitted 2025-05-19 cs.AI cs.LG

Lucas Berry , Axel Brando , Wei-Di Chang , Juan Camilo Gamboa Higuera , David Meger This is my paper · ORCID
classification cs.AIcs.LG
keywords epistemicuncertaintymixtureofexpertstext-to-imagediffusiontraining-freeestimationexpertdisagreementpromptriskmultilingualbiaslatent-spacevariance
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 claims that the disagreement among the expert components of a pre-trained mixture-of-experts text-to-image diffusion model is a usable, training-free signal for how much the model knows about a given prompt. EMoE separates the experts' computation paths from the first denoising step onward, keeps the same initial noise across all paths, and measures the variance of their mid-block latent representations. On COCO and CC3M English prompts, prompts in the lowest uncertainty quartile produce higher text-image alignment scores than prompts in the highest quartile, and the ordering also holds for aesthetic and human-preference metrics. Translated prompts in lower-resource languages produce higher uncertainty and worse alignment, with a correlation of $r = -0.79$ across 25 languages. If the claim is right, one early denoising step can flag risky or under-covered prompts before the expensive generation runs, and can expose language-related training bias.

What carries the argument

The central object is the separated expert path ending in the mid-block latent $m^{\mathrm{post}}_T$. At the first sparse MoE cross-attention layer, EMoE gives each of the $M$ experts its own copy of the noisy latent and the prompt, routes each copy through the rest of the network with the MoE's normal aggregation inside each path, and reads off the $1280 \times 8 \times 8$ mid-block latent after one denoising step. The variance across those $M$ latents, averaged over latent dimensions, is the uncertainty score $EU(y) = \mathbb{E}_{d_{\mathrm{mid}}}\left[\mathrm{Var}_{i \in M}\left(m^{\mathrm{post}}_T\right)\right]$. This mechanism converts a routed mixture into a shared-noise ensemble: same initial noise and same prompt, only expert-specific weights differ, so their divergence is interpreted as disagreement about the prompt. Because the score is available after one step, it can halt the denoising process for high-uncertainty prompts.

What would settle it

Translate a matched set of Finnish prompts that differ only in including a common English loanword such as 'pizza', generate images, and compare EMoE uncertainty with human-rated or ImageReward alignment; if uncertainty drops sharply for loanword prompts while alignment does not improve, the signal is tracking token familiarity rather than prompt-level epistemic uncertainty.

Watch

Extended reading notes

Core claim

The paper's central discovery is that the mean over latent dimensions of the variance across experts at the mid-block of the first denoising step orders prompts by expected text-image alignment. For a prompt $y$, after running the $M$ separated expert paths through one denoising step, the score is $EU(y) = \mathbb{E}_{d_{\mathrm{mid}}}\left[\mathrm{Var}_{i \in M}\left(m^{\mathrm{post}}_T\right)\right]$, where $m^{\mathrm{post}}_T$ is the mid-block latent of path $i$ at the noisiest time step $T$ (the first reverse step). On 40,000 COCO prompts, this score puts prompts into quartiles whose CLIP scores fall from 31.578 to 31.217, with the same monotone pattern for aesthetic score and ImageReward and on CC3M. The same estimator separates English from Finnish translations and yields $r = -0.79$ between mean uncertainty and CLIP score across 25 languages, with languages having more native speakers generally showing lower uncertainty. The paper reads this as evidence that expert disagreement is an estimate of epistemic uncertainty: high variance marks prompts where the model is extrapolating from under-represented training data.

Load-bearing premise

The load-bearing premise is that disagreement among pre-trained fine-tunes of the same base model behaves like independent draws from a posterior over models; if expert divergence mostly reflects style preferences, routing artifacts, or shared-vocabulary effects, the score can still rank prompts without actually measuring epistemic uncertainty.

Editorial extensions

If this is right

  • A text-to-image system can screen prompts before committing to full denoising: high uncertainty means the prompt is likely to produce poorly aligned images, so users can revise or the system can halt early.
  • The score can rank prompts by expected text-image alignment on both COCO and CC3M, so it can serve as a model-specific prompt-quality metric without access to the training data.
  • Model audits can use EMoE to map coverage by language and vocabulary, exposing biases such as lower uncertainty for Finnish prompts that contain the recognizable English loanword 'pizza'.
  • The Fast EMoE variant keeps the early uncertainty check cheap enough for routine use, at about 5.5 seconds per prompt versus 3.58 seconds for standard generation.
  • Because the near-duplicate-checkpoint experiment uses near-identical experts, the method does not require wildly diverse experts to produce a useful uncertainty ordering.

Reading between the lines

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

  • Editorial extension: if inter-expert variance really tracks epistemic uncertainty, it could be used as an acquisition function for active data collection—prompts with high variance are the ones to add to training data, though the paper only demonstrates ranking, not acquisition.
  • Editorial extension: the Finnish 'pizza' result suggests a boundary condition: languages that share vocabulary with English may systematically under-report uncertainty even when their syntax differs, so the score should be validated against human alignment before being used as a fairness audit.
  • Editorial extension: the same first-step variance could be applied to other MoE architectures beyond the two tested checkpoints, but whether it generalises to non-diffusion or non-MoE models is not established by the paper.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 7 minor

Summary. The paper proposes EMoE, a training-free method for estimating epistemic uncertainty in text-to-image diffusion models that contain mixture-of-experts (MoE) layers. The method separates the first sparse MoE layer into multiple expert-specific computation paths, keeps the same initial noise across paths, runs the first denoising step, and computes EU(y) = E_dmid[Var_i(mpost_T)] from the mid-block latents. The authors claim that this quantity ranks prompts by text-image alignment quality on COCO and CC3M more consistently than the DECU baseline, detects out-of-distribution Finnish prompts, reveals language-dependent bias across 25 languages, and can be computed early enough to halt generation. Supporting experiments include quartile-based CLIP/Aesthetic/Image Reward comparisons, multilingual correlations, and ablations over ensemble size, denoising step, latent space, and a second MoE architecture.

Significance. If its central claim holds, EMoE would be a practical and inexpensive diagnostic: it requires no training, no auxiliary networks, and only one early denoising step to produce a prompt-level uncertainty signal, with potential uses in risk assessment and bias auditing of black-box generative models. The paper has real strengths: experiments span COCO and CC3M with multiple metrics, the multilingual analysis is broad, ablations cover the main design choices, and the compute and memory requirements are reported in detail. The central empirical trend is plausible and replicated across datasets and metrics. However, the evidence as presented does not yet separate the proposed signal from a prompt-length proxy, and the interpretation of inter-expert variance as epistemic uncertainty rests on an assumption, not a demonstration.

major comments (4)
  1. [§4.1, Tables 1–2, Appendix G] A second, related issue is the size of the effect. The CLIP difference between Q1 and Q4 is 31.578 vs 31.217, with reported standard deviations of about 0.15–0.16; the Aesthetic and Image Reward gaps are similarly small. With 40,000 prompts, a Jonckheere-Terpstra p-value of 3.34e-19 can be driven by a tiny effect that has little practical utility. The paper should report effect sizes (e.g., rank-biserial correlation, Cohen's d) and, ideally, the per-prompt Spearman correlation between EU and CLIP score, not just quartile means, so that readers can judge whether the relationship is strong enough to support a 'reliable uncertainty-aware prompt signal'.
  2. [§3.2, Eq. (2), Appendix D] The core interpretation of Var_i(mpost_T) as epistemic uncertainty is not established. Appendix D argues that experts can be viewed as i.i.d. samples from a posterior distribution over functions and invokes a Gaussian-process convergence argument, but the proof only shows that an ensemble of i.i.d. functions converges to a GP; it does not show that the four community fine-tunes in the SegMoE model card are i.i.d. draws from a posterior. Those experts are explicitly associated with different style descriptors (photorealistic, anime, fantasy, dreamlike), so inter-expert variance may reflect systematic style preferences or routing artifacts rather than epistemic uncertainty about the prompt. The authors should provide a concrete test that distinguishes these explanations: for example, compare EU against (a) variance across different initial noise seeds on a single expert path, (b) variance induced by style-neutral prompt perturbations of matched length, or (c) the performance of EU on a proper OOD-detection benchmark where length is controlled. If the variance is mostly style or routing signal, the practical ranking claim may partially survive, but the epistemic-uncertainty labeling and the bias-detection conclusions would need to be revised.
  3. [§4.1, Appendix C, Figures 3 and 9] The claim that EMoE ranks prompts 'more consistently than diffusion-specific and router-based baselines' is not backed by a quantitative comparison. The text states that DECU 'did not demonstrate this capability' and the figures annotate 'unlike DECU', but no DECU numbers, rank correlations, AUROC values, or implementation details are reported. Because this contrast appears in the abstract and is load-bearing for the claimed advantage over existing uncertainty methods, the authors should include direct quantitative comparisons: same prompts, same metrics, and a clear description of how DECU was configured. A router-only baseline (e.g., using the gating weights without expert separation) would also be valuable to show that the signal comes from disagreement rather than from routing behavior.
  4. [§4.2 and §4.3, Tables 3 and 8, Figures 6 and 8] The multilingual bias conclusions are drawn from language-level aggregates. The reported correlation r=-0.79 (CLIP) and r=-0.73 (Image Reward) is computed across language mean points; averaging within language can inflate correlation relative to the per-prompt relationship, and no confidence intervals or per-prompt correlations are given. The claim of a 'strong negative correlation' needs to be supported by per-prompt analysis or a mixed-effects model with language as a random effect. In addition, the paper states that CLIP scores for non-English prompts are evaluated using the English version of the prompt, which means the CLIP model is effectively scoring English-text-to-image alignment even for Finnish or other languages. This introduces a potential language-specific measurement bias: differences in CLIP score may reflect CLIP's own English bias rather than generation quality alone. The authors should state this limitation explicitly in the main text and consider a language-matched evaluation or a secondary metric that is less language-sensitive.
minor comments (7)
  1. [Appendix C] The text says the CC3M results are 'presented in Figure 9 and Table 6', but Table 6 in the main text is the compute-requirements table; the reference should be to Table 9.
  2. [Equation (2) and Appendix A] Equation (2) defines EU(y) as E_dmid[Var_i(mpost_T)], but Appendix A states that the reported uncertainty is sqrt(dmidsize) × EU(y). Please reconcile the scaling and define which quantity is used in all tables and figures.
  3. [§3.1] The description of the separated paths is ambiguous: after the first layer is separated, later sparse MoE layers are said to aggregate outputs 'within each computational path according to Equation 1'. Please clarify how the gate weights and selected expert sets are computed when the first-layer routing has been replaced by an explicit separation, and whether the same gate weights are used across all paths.
  4. [Abstract and §4] The abstract promises comparison against 'diffusion-specific and router-based baselines', but the only baseline discussed in the evaluation is DECU. Either add a router-based baseline or soften the claim to match what is actually compared.
  5. [Figure 7a] The ablation on ensemble size says ensembles of sizes 2 and 3 are 'sub-optimal' because Q1 yields a lower CLIP score than Q2, but no statistical significance or confidence intervals are provided. Please add error bars or significance tests, or phrase the conclusion more cautiously.
  6. [§3.2 and Algorithm 1] The text says uncertainty is computed 'after the first denoising step', while Algorithm 1 extracts mpost_T for t = T at the initial step. Please make the notation consistent so that the reported step index matches the algorithm.
  7. [Figure 1] The example reports uncertainty values of 0.38 and 0.83, while Table 3 reports mean uncertainties of 0.37 ± 0.14 and 0.48 ± 0.19 for English and Finnish. The Finnish example is far above the mean; please note that it is a selected extreme example rather than a typical case.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: EMoE's uncertainty estimate is a fixed training-free function of pre-trained expert activations, validated against external CLIP/Image-Reward benchmarks rather than derived from them.

full rationale

The derivation chain is not circular in the sense defined here. EU(y) in Eq. (2) is a direct, parameter-free statistic of the mid-block latents mpost_T from M separated expert paths; no parameter is fitted to CLIP score, Image Reward, or any target quality metric, and no quantity is renamed as a prediction after being used as an input. The identification of inter-expert variance with epistemic uncertainty is an operational assumption grounded in the standard ensemble literature (e.g., Lakshminarayanan et al.), not a result derived from the validation data; Appendix D's GP discussion is a loose intuitive justification, not a theorem used to force the estimator. The main evidence, the monotone CLIP/Aesthetic/Image-Reward trends across uncertainty quartiles on COCO and CC3M and the negative multilingual correlation, is an external check rather than a construction-level equivalence. The paper does cite the authors' prior uncertainty work (DECU, Berry-Meger) for context and baselines, but those citations are not load-bearing: EMoE's validity does not depend on a uniqueness theorem or on an unverified self-citation, and DECU is used as a comparison point that EMoE outperforms. The only mild methodological concern, that the latent-space choice (mpost vs mpre vs zT-1) and the first-step choice were selected after inspecting the same COCO quartile results, is a selection-on-validation issue, not a circular reduction of Eq. (2) to the target metric; the trend also replicates on CC3M, Finnish, and 25-language evaluations. Accordingly, no specific circular step meets the quoted-equation standard required for a positive finding.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

No new physical or architectural entities are introduced; the expert computation paths are a decomposition of existing MoE layers. The only fitted quantity is the ensemble size M=4, chosen by an ablation on the same COCO metric. The core epistemic-uncertainty interpretation rests on the ensemble assumption listed above, not on a derived or externally anchored definition.

free parameters (1)
  • Number of experts M = 4
    Set to 4 in Section 3.3 and Appendix F; the ablation in Figure 7a selects 4 because ensemble sizes 2 and 3 give suboptimal Q1 versus Q2 CLIP ordering on the COCO evaluation set.
assumptions (5)
  • ad hoc to paper Pre-trained experts in a MoE diffusion model form an ensemble whose inter-expert variance is epistemic uncertainty.
    Central assumption of Section 3.2 and Equation 2; Appendix D attempts a GP justification but assumes i.i.d. weights that the merged community checkpoints do not satisfy.
  • domain assumption CLIP score is a valid proxy for text-image alignment.
    Used as the primary evaluation metric throughout; Appendix B acknowledges CLIP biases but does not quantify their effect on the quartile comparisons.
  • domain assumption Machine-translated prompts are a valid out-of-distribution perturbation for studying language coverage.
    Section 4.2 uses Google Translate without human verification; shared-vocabulary effects are mentioned but not controlled.
  • domain assumption The mid-block latent after the first denoising step captures prompt-dependent semantic variation.
    Section 3.2 chooses mpost_T as the measurement space; the ablation in Figure 7c compares alternatives but does not establish that mpost_T is semantically interpretable.
  • domain assumption COCO prompts are in-distribution for the pre-trained experts.
    Section 4.1 asserts 'each expert has been trained on all data in the COCO dataset' without evidence; the cited base models are typically trained on LAION, not COCO, so the in-distribution claim is unsupported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of EMoE: Training-Free Expert Disagreement for Uncertainty-Aware Text-to-Image Diffusion." pith.science (2026). https://pith.science/paper/KYFG42MW

@misc{pith2026250513273,
  author       = {Pith},
  title        = {Pith review of: EMoE: Training-Free Expert Disagreement for Uncertainty-Aware Text-to-Image Diffusion},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KYFG42MW}},
  note         = {Machine review of arXiv:2505.13273}
}
read the original abstract

Large text-to-image diffusion models rarely expose reliable signals of when a prompt is likely to produce a poorly aligned generation, especially when training data is undisclosed. We study whether expert disagreement inside pre-trained mixture-of-experts (MoE) diffusion models can serve as a reliable estimate for epistemic uncertainty. We introduce EMoE, a training-free method that separates expert-specific computation paths at an early MoE layer, uses the same initial noise across paths, and measures variance among their latent representations after the first denoising step. This provides an uncertainty-aware prompt signal before full image generation, without auxiliary networks or training diffusion ensembles. On COCO and CC3M, EMoE ranks prompts by text-image alignment quality metrics more consistently than diffusion-specific and router-based baselines. We further apply EMoE to multilingual prompts and find systematic language-dependent differences in disagreement and generation quality, including shared-vocabulary effects. These results position EMoE as a practical diagnostic tool for prompt risk, model coverage, and bias analysis in MoE text-to-image diffusion models.

Figures

Figures reproduced from arXiv: 2505.13273 by the authors.

Figure 1
Figure 1. This figure presents the model’s uncertainty estimates when interpreting the same prompt [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. EMoE separates expert components in the first cross-attention layer in the first [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. CLIP score across different uncertainty quartiles. EMoE accurately attributes prompts that produce images with high CLIP scores with low uncertainty unlike DECU. The red line indicates the average CLIP score across all quartiles [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Uncertainty distribution for Finnish and English prompts, showing higher uncertainty for Finnish prompts compared to English [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Qualitative comparison of image-generation for a Finnish prompt with the word “pizza” [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Negative correlation between uncertainty [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Ablation studies validating EMoE hyperparameters: ensemble size (a), denoising step (b), [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]
Figure 8
Figure 8. Figure 8: Negative correlation between un￾certainty and image quality across prompts translated into 25 different languages. EMoE demonstrates a strong negative correlation (r = -0.73) between uncertainty and Image Reward, with languages having more native speakers generally pro…
Figure 9
Figure 9. Figure 9: CLIP score on the CC3M dataset across different uncertainty quartiles. EMoE ac￾curately attributes prompts that produce images with high CLIP scores with low uncertainty un￾like DECU. The red line indicates the average CLIP score across all quartiles [PITH_FULL_IMAGE:…
Figure 10
Figure 10. Figure 10: This pictures depicts how to have accurate gates without training. Each expert is associated with a positive and a neg￾ative descriptor, ψ i = [PITH_FULL_IMAGE:figures/full_fig_p019_10.png]
Figure 11
Figure 11. Figure 11: EMoE’s uncertainty across different prompts: Each row represents a distinct prompt, [PITH_FULL_IMAGE:figures/full_fig_p022_11.png]
Figure 12
Figure 12. Figure 12: Qualitative comparison of image-generation for a Finnish prompt with the word “pizza" [PITH_FULL_IMAGE:figures/full_fig_p022_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

68 extracted references · 14 canonical work pages

  1. [1]

    Clip the bias: How useful is balancing data in multimodal learning? arXiv preprint arXiv:2403.04547, 2024

    Ibrahim Alabdulmohsin, Xiao Wang, Andreas Steiner, Priya Goyal, Alexander D’Amour, and Xiaohua Zhai. Clip the bias: How useful is balancing data in multimodal learning? arXiv preprint arXiv:2403.04547, 2024

    arXiv 2024

  2. [2]

    ediff-i: Text-to-image diffusion models with an ensemble of expert denoisers

    Yogesh Balaji, Seungjun Nah, Xun Huang, Arash Vahdat, Jiaming Song, Karsten Kreis, Miika Aittala, Timo Aila, Samuli Laine, Bryan Catanzaro, et al. ediff-i: Text-to-image diffusion models with an ensemble of expert denoisers. arXiv preprint arXiv:2211.01324, 2022

    arXiv 2022

  3. [3]

    Lumiere: A space-time diffusion model for video generation

    Omer Bar-Tal, Hila Chefer, Omer Tov, Charles Herrmann, Roni Paiss, Shiran Zada, Ariel Ephrat, Junhwa Hur, Yuanzhen Li, Tomer Michaeli, et al. Lumiere: A space-time diffusion model for video generation. arXiv preprint arXiv:2401.12945, 2024

    arXiv 2024

  4. [4]

    Efficient epistemic uncertainty estimation in regression ensemble models using pairwise-distance estimators

    Lucas Berry and David Meger. Efficient epistemic uncertainty estimation in regression ensemble models using pairwise-distance estimators. arXiv preprint arXiv:2308.13498, 2023

    arXiv 2023

  5. [5]

    Normalizing flow ensembles for rich aleatoric and epistemic uncertainty modeling

    Lucas Berry and David Meger. Normalizing flow ensembles for rich aleatoric and epistemic uncertainty modeling. Proceedings of the AAAI Conference on Artificial Intelligence, 37(6): 6806–6814, 2023

    work page 2023

  6. [6]

    Shedding light on large generative networks: Estimating epistemic uncertainty in diffusion models

    Lucas Berry, Axel Brando, and David Meger. Shedding light on large generative networks: Estimating epistemic uncertainty in diffusion models. In The 40th Conference on Uncertainty in Artificial Intelligence, 2024

    work page 2024

  7. [7]

    Random forests

    Leo Breiman. Random forests. Machine learning, 45:5–32, 2001

    2001

  8. [8]

    Hyper-diffusion: Estimating epistemic and aleatoric uncertainty with a single model

    Matthew A Chan, Maria J Molina, and Christopher A Metzler. Hyper-diffusion: Estimating epistemic and aleatoric uncertainty with a single model. arXiv preprint arXiv:2402.03478, 2024

    arXiv 2024

Show all 68 references
  1. [9]

    Tibet: Identifying and evaluating biases in text-to-image generative models

    Aditya Chinchure, Pushkar Shukla, Gaurav Bhatt, Kiri Salij, Kartik Hosanagar, Leonid Sigal, and Matthew Turk. Tibet: Identifying and evaluating biases in text-to-image generative models. arXiv preprint arXiv:2312.01261, 2023

    2023 arXiv

  2. [10]

    Waic, but why? generative ensembles for robust anomaly detection

    Hyunsun Choi, Eric Jang, and Alexander A Alemi. Waic, but why? generative ensembles for robust anomaly detection. arXiv preprint arXiv:1810.01392, 2018

    2018 arXiv

  3. [11]

    Deep reinforcement learning in a handful of trials using probabilistic dynamics models

    Kurtland Chua, Roberto Calandra, Rowan McAllister, and Sergey Levine. Deep reinforcement learning in a handful of trials using probabilistic dynamics models. In Advances in Neural Information Processing Systems, volume 31, 2018

    2018

  4. [12]

    Decomposition of uncertainty in bayesian deep learning for efficient and risk-sensitive learning

    Stefan Depeweg, Jose-Miguel Hernandez-Lobato, Finale Doshi-Velez, and Steffen Udluft. Decomposition of uncertainty in bayesian deep learning for efficient and risk-sensitive learning. In International Conference on Machine Learning, pages 1184–1193. PMLR, 2018

    2018

  5. [13]

    Aleatory or epistemic? does it matter? Structural safety, 31(2):105–112, 2009

    Armen Der Kiureghian and Ove Ditlevsen. Aleatory or epistemic? does it matter? Structural safety, 31(2):105–112, 2009

    2009

  6. [14]

    Diffusion models beat gans on image synthesis

    Prafulla Dhariwal and Alexander Nichol. Diffusion models beat gans on image synthesis. Advances in neural information processing systems, 34:8780–8794, 2021

    2021

  7. [15]

    Uncertainty quantification for deep unrolling-based computational imaging

    Canberk Ekmekci and Mujdat Cetin. Uncertainty quantification for deep unrolling-based computational imaging. IEEE Transactions on Computational Imaging, 8:1195–1209, 2022. 10

    2022

  8. [16]

    Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity

    William Fedus, Barret Zoph, and Noam Shazeer. Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity. Journal of Machine Learning Research, 23 (120):1–39, 2022

    2022

  9. [17]

    Addressing function approximation error in actor-critic methods

    Scott Fujimoto, Herke Hoof, and David Meger. Addressing function approximation error in actor-critic methods. In International conference on machine learning, pages 1587–1596. PMLR, 2018

    2018

  10. [18]

    Deep bayesian active learning with image data

    Yarin Gal, Riashat Islam, and Zoubin Ghahramani. Deep bayesian active learning with image data. In International Conference on Machine Learning, pages 1183–1192. PMLR, 2017

    2017

  11. [19]

    Arcee’s mergekit: A toolkit for merging large language models

    Charles Goddard, Shamane Siriwardhana, Malikeh Ehghaghi, Luke Meyers, Vlad Karpukhin, Brian Benedict, Mark McQuade, and Jacob Solawetz. Arcee’s mergekit: A toolkit for merging large language models. arXiv preprint arXiv:2403.13257, 2024

    2024 arXiv

  12. [20]

    Sources of uncertainty in machine learning–a statisticians’ view

    Cornelia Gruber, Patrick Oliver Schenk, Malte Schierholz, Frauke Kreuter, and Göran Kauer- mann. Sources of uncertainty in machine learning–a statisticians’ view. arXiv preprint arXiv:2305.16703, 2023

    2023 arXiv

  13. [21]

    Towards the systematic reporting of the energy and carbon footprints of machine learning

    Peter Henderson, Jieru Hu, Joshua Romoff, Emma Brunskill, Dan Jurafsky, and Joelle Pineau. Towards the systematic reporting of the energy and carbon footprints of machine learning. Journal of Machine Learning Research, 21(248):1–43, 2020

    2020

  14. [22]

    Clipscore: A reference-free evaluation metric for image captioning

    Jack Hessel, Ari Holtzman, Maxwell Forbes, Ronan Le Bras, and Yejin Choi. Clipscore: A reference-free evaluation metric for image captioning. arXiv preprint arXiv:2104.08718, 2021

    2021 arXiv

  15. [23]

    Denoising diffusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models. Advances in Neural Information Processing Systems, 33:6840–6851, 2020

    2020

  16. [24]

    Imagen video: High definition video generation with diffusion models

    Jonathan Ho, William Chan, Chitwan Saharia, Jay Whang, Ruiqi Gao, Alexey Gritsenko, Diederik P Kingma, Ben Poole, Mohammad Norouzi, David J Fleet, et al. Imagen video: High definition video generation with diffusion models. arXiv preprint arXiv:2210.02303, 2022

    arXiv

  17. [25]

    Video diffusion models

    Jonathan Ho, Tim Salimans, Alexey Gritsenko, William Chan, Mohammad Norouzi, and David J Fleet. Video diffusion models. Advances in Neural Information Processing Systems, 35:8633–8646, 2022

    2022

  18. [26]

    Deep ensembles from a bayesian perspective

    Lara Hoffmann and Clemens Elster. Deep ensembles from a bayesian perspective. arXiv preprint arXiv:2105.13283, 2021

    2021 arXiv

  19. [27]

    Aleatory and epistemic uncertainty in probability elicitation with an example from hazardous waste management

    Stephen C Hora. Aleatory and epistemic uncertainty in probability elicitation with an example from hazardous waste management. Reliability Engineering & System Safety, 54(2-3):217–223, 1996

    1996

  20. [28]

    Prodiff: Progressive fast diffusion model for high-quality text-to-speech

    Rongjie Huang, Zhou Zhao, Huadai Liu, Jinglin Liu, Chenye Cui, and Yi Ren. Prodiff: Progressive fast diffusion model for high-quality text-to-speech. In Proceedings of the 30th ACM International Conference on Multimedia, pages 2595–2605, 2022

    2022

  21. [29]

    Aleatoric and epistemic uncertainty in machine learning: An introduction to concepts and methods

    Eyke Hüllermeier and Willem Waegeman. Aleatoric and epistemic uncertainty in machine learning: An introduction to concepts and methods. Machine Learning, 110(3):457–506, 2021

    2021

  22. [30]

    Image-to-image translation with conditional adversarial networks

    Phillip Isola, Jun-Yan Zhu, Tinghui Zhou, and Alexei A Efros. Image-to-image translation with conditional adversarial networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 1125–1134, 2017

    2017

  23. [31]

    Adaptive mixtures of local experts

    Robert A Jacobs, Michael I Jordan, Steven J Nowlan, and Geoffrey E Hinton. Adaptive mixtures of local experts. Neural computation, 3(1):79–87, 1991

    1991

  24. [32]

    Imagic: Text-based real image editing with diffusion models

    Bahjat Kawar, Shiran Zada, Oran Lang, Omer Tov, Huiwen Chang, Tali Dekel, Inbar Mosseri, and Michal Irani. Imagic: Text-based real image editing with diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 6007–6017, 2023

    2023

  25. [33]

    What uncertainties do we need in bayesian deep learning for computer vision? Advances in neural information processing systems, 30, 2017

    Alex Kendall and Yarin Gal. What uncertainties do we need in bayesian deep learning for computer vision? Advances in neural information processing systems, 30, 2017. 11

    2017

  26. [34]

    Text2video-zero: Text-to-image diffusion models are zero-shot video generators

    Levon Khachatryan, Andranik Movsisyan, Vahram Tadevosyan, Roberto Henschel, Zhangyang Wang, Shant Navasardyan, and Humphrey Shi. Text2video-zero: Text-to-image diffusion models are zero-shot video generators. In Proceedings of the IEEE/CVF International Conference on Computer ...

    2023

  27. [35]

    Auto-encoding variational bayes

    Diederik P Kingma and Max Welling. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114, 2013

    2013 arXiv

  28. [36]

    Batchbald: Efficient and diverse batch acquisition for deep bayesian active learning

    Andreas Kirsch, Joost Van Amersfoort, and Yarin Gal. Batchbald: Efficient and diverse batch acquisition for deep bayesian active learning. Advances in neural information processing systems, 32, 2019

    2019

  29. [37]

    Multi- concept customization of text-to-image diffusion

    Nupur Kumari, Bingliang Zhang, Richard Zhang, Eli Shechtman, and Jun-Yan Zhu. Multi- concept customization of text-to-image diffusion. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1931–1941, 2023

    1931

  30. [38]

    Simple and scalable predictive uncertainty estimation using deep ensembles

    Balaji Lakshminarayanan, Alexander Pritzel, and Charles Blundell. Simple and scalable predictive uncertainty estimation using deep ensembles. Advances in neural information processing systems, 30, 2017

    2017

  31. [39]

    Gshard: Scaling giant models with condi- tional computation and automatic sharding

    Dmitry Lepikhin, HyoukJoong Lee, Yuanzhong Xu, Dehao Chen, Orhan Firat, Yanping Huang, Maxim Krikun, Noam Shazeer, and Zhifeng Chen. Gshard: Scaling giant models with condi- tional computation and automatic sharding. arXiv preprint arXiv:2006.16668, 2020

    2006 arXiv

  32. [40]

    Microsoft coco: Common objects in context

    Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dollár, and C Lawrence Zitnick. Microsoft coco: Common objects in context. In Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings...

    2014

  33. [41]

    Uncertainty estimation and quantification for llms: A simple supervised approach

    Linyu Liu, Yu Pan, Xiaocheng Li, and Guanting Chen. Uncertainty estimation and quantification for llms: A simple supervised approach. arXiv preprint arXiv:2404.15993, 2024

    2024 arXiv

  34. [42]

    Instaflow: One step is enough for high-quality diffusion-based text-to-image generation

    Xingchao Liu, Xiwen Zhang, Jianzhu Ma, Jian Peng, et al. Instaflow: One step is enough for high-quality diffusion-based text-to-image generation. In The Twelfth International Conference on Learning Representations, 2023

    2023

  35. [43]

    Training of neural networks with uncertain data, a mixture of experts approach

    Lucas Luttner. Training of neural networks with uncertain data, a mixture of experts approach. arXiv preprint arXiv:2312.08083, 2023

    2023 arXiv

  36. [44]

    Uncertainty estimation in autoregressive structured prediction

    Andrey Malinin and Mark Gales. Uncertainty estimation in autoregressive structured prediction. arXiv preprint arXiv:2002.07650, 2020

    2002 arXiv

  37. [45]

    Glide: Towards photorealistic image genera- tion and editing with text-guided diffusion models

    Alexander Quinn Nichol, Prafulla Dhariwal, Aditya Ramesh, Pranav Shyam, Pamela Mishkin, Bob Mcgrew, Ilya Sutskever, and Mark Chen. Glide: Towards photorealistic image genera- tion and editing with text-guided diffusion models. In International Conference on Machine Learning, p...

    2022

  38. [46]

    Deep exploration via bootstrapped dqn

    Ian Osband, Charles Blundell, Alexander Pritzel, and Benjamin Van Roy. Deep exploration via bootstrapped dqn. Advances in neural information processing systems, 29, 2016

    2016

  39. [47]

    The hidden uncertainty in a neural networks activations.arXiv preprint arXiv:2012.03082, 2020

    Janis Postels, Hermann Blum, Yannick Strümpler, Cesar Cadena, Roland Siegwart, Luc Van Gool, and Federico Tombari. The hidden uncertainty in a neural networks activations.arXiv preprint arXiv:2012.03082, 2020

    2012 arXiv

  40. [48]

    Hierarchical text-conditional image generation with clip latents

    Aditya Ramesh, Prafulla Dhariwal, Alex Nichol, Casey Chu, and Mark Chen. Hierarchical text-conditional image generation with clip latents. arXiv preprint arXiv:2204.06125, 1(2):3, 2022

    2022 arXiv

  41. [49]

    High- resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. High- resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 10684–10695, 2022. 12

    2022

  42. [50]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. In Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part ...

    2015

  43. [51]

    Photorealistic text-to-image diffusion 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 diffusion models with deep language understanding. Advances in Neural Information...

    2022

  44. [52]

    Laion- 5b: An open large-scale dataset for training next generation image-text models

    Christoph Schuhmann, Romain Beaumont, Richard Vencu, Cade Gordon, Ross Wightman, Mehdi Cherti, Theo Coombes, Aarush Katta, Clayton Mullis, Mitchell Wortsman, et al. Laion- 5b: An open large-scale dataset for training next generation image-text models. Advances in Neural Inform...

    2022

  45. [53]

    Conceptual captions: A cleaned, hypernymed, image alt-text dataset for automatic image captioning

    Piyush Sharma, Nan Ding, Sebastian Goodman, and Radu Soricut. Conceptual captions: A cleaned, hypernymed, image alt-text dataset for automatic image captioning. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ...

    2018

  46. [54]

    Outrageously large neural networks: The sparsely-gated mixture-of-experts layer

    Noam Shazeer, Azalia Mirhoseini, Krzysztof Maziarz, Andy Davis, Quoc Le, Geoffrey Hinton, and Jeff Dean. Outrageously large neural networks: The sparsely-gated mixture-of-experts layer. arXiv preprint arXiv:1701.06538, 2017

    2017 arXiv

  47. [55]

    Improving image captioning with better use of captions

    Zhan Shi, Xu Zhou, Xipeng Qiu, and Xiaodan Zhu. Improving image captioning with better use of captions. arXiv preprint arXiv:2006.11807, 2020

    2006 arXiv

  48. [56]

    Deep unsuper- vised learning using nonequilibrium thermodynamics

    Jascha Sohl-Dickstein, Eric Weiss, Niru Maheswaranathan, and Surya Ganguli. Deep unsuper- vised learning using nonequilibrium thermodynamics. In International Conference on Machine Learning, pages 2256–2265. PMLR, 2015

    2015

  49. [57]

    Denoising diffusion implicit models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denoising diffusion implicit models. In International Conference on Learning Representations, 2020

    2020

  50. [58]

    High frequency matters: Uncertainty guided image compression with wavelet diffusion

    Juan Song, Jiaxiang He, Mingtao Feng, Keyan Wang, Yunsong Li, and Ajmal Mian. High frequency matters: Uncertainty guided image compression with wavelet diffusion. arXiv preprint arXiv:2407.12538, 2024

    2024

  51. [59]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, et al. Attention is all you need. Advances in neural information processing systems , 30(1): 261–272, 2017

    2017

  52. [60]

    Reducing llm hallucinations using epistemic neural networks

    Shreyas Verma, Kien Tran, Yusuf Ali, and Guangyu Min. Reducing llm hallucinations using epistemic neural networks. arXiv preprint arXiv:2312.15576, 2023

    2023 arXiv

  53. [61]

    Diffusers: State-of-the-art diffusion models

    Patrick von Platen, Suraj Patil, Anton Lozhkov, Pedro Cuenca, Nathan Lambert, Kashif Rasul, Mishig Davaadorj, Dhruv Nair, Sayak Paul, William Berman, Yiyi Xu, Steven Liu, and Thomas Wolf. Diffusers: State-of-the-art diffusion models. https://github.com/huggingface/ diffusers, 2022

    2022

  54. [62]

    Epistemic uncertainty quantification for pre-trained neural networks

    Hanjing Wang and Qiang Ji. Epistemic uncertainty quantification for pre-trained neural networks. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 11052–11061, 2024

    2024

  55. [63]

    Fast diffusion model

    Zike Wu, Pan Zhou, Kenji Kawaguchi, and Hanwang Zhang. Fast diffusion model. arXiv preprint arXiv:2306.06991, 2023

    2023 arXiv

  56. [64]

    Imagereward: Learning and evaluating human preferences for text-to-image generation

    Jiazheng Xu, Xiao Liu, Yuchen Wu, Yuxuan Tong, Qinkai Li, Ming Ding, Jie Tang, and Yuxiao Dong. Imagereward: Learning and evaluating human preferences for text-to-image generation. Advances in Neural Information Processing Systems, 36, 2024

    2024

  57. [65]

    Segmoe: Segmind mixture of diffusion experts

    Harish Prabhala Yatharth Gupta, Vishnu V Jaddipal. Segmoe: Segmind mixture of diffusion experts. https://github.com/segmind/segmoe, 2024. 13

    2024

  58. [66]

    One-step diffusion with distribution matching distillation

    Tianwei Yin, Michaël Gharbi, Richard Zhang, Eli Shechtman, Frédo Durand, William T Freeman, and Taesung Park. One-step diffusion with distribution matching distillation. In CVPR, 2024

    2024

  59. [67]

    Efficient deweahter mixture-of-experts with uncertainty-aware feature-wise linear modulation

    Rongyu Zhang, Yulin Luo, Jiaming Liu, Huanrui Yang, Zhen Dong, Denis Gudovskiy, Tomoyuki Okuno, Yohei Nakata, Kurt Keutzer, Yuan Du, et al. Efficient deweahter mixture-of-experts with uncertainty-aware feature-wise linear modulation. In Proceedings of the AAAI Conference on Ar...

    2024

  60. [68]

    cinematic , portrait , photograph , instagram , fashion , movie , macro shot , 8K , RAW , hyperrealistic , ultra realistic ,

    Zhuobin Zheng, Chun Yuan, Xinrui Zhu, Zhihui Lin, Yangyang Cheng, Cheng Shi, and Jiahui Ye. Self-supervised mixture-of-experts by uncertainty estimation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 33, pages 5933–5940, 2019. 14 A Compute Details Ta...

    2019

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.