Pith. sign in

REVIEW 4 major objections 6 minor 39 references

Diffusion Models with Adaptive Negative Sampling Without External Resources

T0 review · 4 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read The paper proposes ANSWER, a training-free sampling procedure that recomputes a negative-guidance noise direction at every diffusion step inside the latent space, claiming it improves prompt adherence and human preference over…

desk verdict A clearly-specified training-free negative-sampling upgrade with real but unproven adaptivity; the evaluation is too thin to support the headline claim as-is. read the letter →

arxiv 2508.02973 v1 pith:DLFNRTOF submitted 2025-08-05 cs.CV

Alakh Desai , Nuno Vasconcelos This is my paper
classification cs.CV
keywords adaptivenegativesamplingdiffusionmodelsclassifier-freeguidancepromptingtraining-freelatent-spacepromptadherence
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

The paper proposes ANSWER, a sampling procedure for text-to-image diffusion models that computes a negative-guidance noise direction anew at each denoising step, entirely inside the model's latent space. Its starting point is the observation that the optimal negative prompt for a given positive prompt changes as the noisy image evolves along the diffusion chain, so any single negative text prompt, whether hand-written or produced by a captioning model, is only optimal at one instant. ANSWER replaces the fixed negative-condition noise of negative prompting with the output of a short diffusion-negative-sampling chain started from the current latent, normalized to match the current noise statistics, and then applies the standard negative-prompting combination. The paper claims that this training-free, external-resource-free change improves prompt adherence, image quality, and human preference over classifier-free guidance and over diffusion-negative prompting on five benchmark suites. If correct, ANSWER is a drop-in improvement for any diffusion model that supports classifier-free guidance.

What carries the argument

The central object is the diffusion-negative-sampling (DNS) chain from prior work, defined by $\hat{\epsilon} = \hat{\epsilon}_\phi + s_n(\hat{\epsilon}_\phi - \hat{\epsilon}_p)$, which generates "negative images" and, in DNP, is captioned to obtain a negative prompt. ANSWER repurposes this chain as a per-step oracle: instead of completing the chain and captioning it, it truncates the chain after $K_t$ steps and takes the noise $\hat{\epsilon}_{t_n}$ at time $t-K_t$ as a substitute for the negative-condition noise in the negative-prompting equation. Two auxiliary mechanisms carry the argument: a scheduler $K_t = K \cdot \frac{2t-T}{2t-T+20} \cdot \frac{T+20}{T}$ for $t \in (T, ..., T/2)$ that smoothly transitions the chain from negative to unconditional guidance, and a normalization step that rescales $\hat{\epsilon}_{t_n}$ to the mean and variance of the unconditional noise $\hat{\epsilon}_\phi$ at time $t$, which the ablations show removes blur introduced by the $K$-step gap between the two chains.

What would settle it

Run ANSWER on a set of prompts while replacing the per-step negative noise with either the DNS noise computed once at the first step and reused everywhere, or with the normalized noise of an unrelated prompt; if prompt-adherence metrics stay roughly constant, then the adaptive component of ANSWER is not doing the work, and only the stronger guidance direction from the DNS chain matters. The paper's own $K$ ablation shows the trade-off, but it does not isolate adaptivity from the mere presence of a DNS-derived negative direction.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central claim is that negative guidance can be made fully automatic and adaptive by replacing the fixed negative prompt in negative prompting with a negative noise estimate produced on the fly: at step $t$, run $K_t$ iterations of the diffusion-negative sampling chain, $\hat{\epsilon} = \hat{\epsilon}_\phi + s_n(\hat{\epsilon}_\phi - \hat{\epsilon}_p)$, initialized at the current latent $z_t$, and use the resulting noise at time $t-K_t$ as $\hat{\epsilon}_n$ in the negative-prompting equation $\hat{\epsilon} = \hat{\epsilon}_n + s(\hat{\epsilon}_p - \hat{\epsilon}_n)$. The paper motivates this by arguing that the ideal negative prompt $n^*(p,t)$ defined through the odds ratio $p_\theta(p|z_t)/p_\theta(n|z_t)$ varies with $t$, and that DNS provides the direction $-\hat{\epsilon}_p$ that maximizes this ratio. It then shows that a $K_t = 1$ version collapses to classifier-free guidance with a rescaled scale, so the method's benefit rests on $K_t > 1$, a hand-designed scheduler that decays $K_t$ over the first half of the chain, and a normalization that matches the mean and variance of the negative noise to the unconditional noise. The experiments report that SDXL + ANSWER outperforms SDXL with CFG and SDXL with DNP on CLIP score, ImageReward, HPSv2, and PickScore across the Attend&Excite, Pick-a-Pic, DrawBench, PartiPrompts, and ImageNet sets, with win rates around 50% and a 2x human preference over the baselines in a human evaluation study.

Load-bearing premise

The method assumes that a short DNS chain started from the current latent $z_t$ yields a noise estimate that faithfully represents the optimal negative direction $n^*(p,t)$ for that step, so that using it in the negative-prompting equation improves the odds of the positive prompt; the paper only proves the $K_t=1$ degenerate case and relies on empirical ablations for $K_t>1$.

Editorial extensions

If this is right

  • Any diffusion model that already uses classifier-free guidance can switch to ANSWER without retraining or extra text inputs.
  • Per-step negative noise removes the two main costs of diffusion-negative prompting: the external captioning model and the fixed one-time negative caption.
  • For prompts where high CFG scale fails, such as spatial relations, numeracy, and imaginary scenes, ANSWER's odds-ratio boosting can force compliance, as the paper's qualitative examples show.
  • Human preference studies reported in the paper put ANSWER ahead of both CFG and DNP by roughly 2-to-1, with win rates of 41–52% on automated metrics.
  • The method's gains are largest in the early denoising steps; after the midpoint, CFG suffices, which bounds the extra inference cost to $O(KT)$ with small $K$.

Reading between the lines

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

  • The paper fixes $K=5$ for all prompts; a prompt- and seed-dependent policy for $K$ or for the stopping time $t=T/2$ is a natural next step, since the paper's own ablation shows quality degrades when $K$ is too large and different prompts need different correction strength.
  • Because ANSWER operates entirely on noise estimates, it is composable with attention-based latent editing methods; the combination could fix spatial binding and anatomy simultaneously, though the paper does not test this.
  • The claim that the ideal negative prompt shifts along the chain could be tested directly by comparing DNS images at consecutive steps with caption similarity; if the negatives only change cosmetically, a single negative would suffice and ANSWER's adaptivity would be unnecessary.
  • An information-theoretic reading is that ANSWER distills a per-step negative text prompt into a noise vector, avoiding the quantization loss of captioning; this suggests the method may be most valuable for fine-grained attributes that captioners routinely miss, which the paper observes qualitatively.
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

4 major / 6 minor

Summary. The paper proposes ANSWER (Adaptive Negative Sampling Without External Resources), a training-free modification of classifier-free guidance (CFG) sampling for text-to-image diffusion models. At each early denoising step, ANSWER runs a short K_t-step chain of Diffusion Negative Sampling (DNS), starting from the current latent, to produce a negative noise estimate that replaces the unconditional noise in the negative-prompting update of Eq. (5). The negative noise is normalized to match the statistics of the unconditional noise, and a scheduler decays K_t as t decreases. The authors argue that the optimal negative prompt n*(p,t) changes with the diffusion step, and that ANSWER approximates it directly in noise space, avoiding external captioning. Experiments on SDXL and SD compare ANSWER against CFG and DNP across Attend&Excite, Pick-a-Pic, DrawBench, PartiPrompts, and ImageNet using CLIP score, IS, FID, HPSv2, ImageReward, PickScore, plus an AMT human study. The paper reports that ANSWER outperforms the baselines on most metrics and is preferred by humans roughly two to one.

Significance. If the central claim is correct, ANSWER is a practically valuable drop-in sampler: it requires no training, no external captioner, and no user-specified negative prompt, and it can be applied to any model that supports CFG. The method is specified in full in Algorithm 1, which is a real strength: the procedure is reproducible in principle, and the ablations on K and on the normalization step are useful. The paper also addresses a genuine limitation of prior DNP work, namely that a single negative prompt estimated once cannot adapt to the changing latent along the diffusion chain. However, the significance is not yet established. The key mechanism—that a K_t-step DNS chain produces a noise estimate that acts as the optimal negative direction at step t—is asserted rather than derived or directly measured for K_t>1. The empirical evidence lacks error bars and significance tests, and one reported metric contradicts the summary text. The contribution is promising and within scope, but the load-bearing 'adaptive' claim needs a control experiment that isolates per-step recomputation.

major comments (4)
  1. The load-bearing step of ANSWER is the replacement of the negative noise term in Eq. (5) with the noise produced by a K_t-step DNS chain run from the current latent z_t. The paper only analyzes the K_t=1 limit, where the construction reduces to CFG with rescaled guidance, and for K_t>1 it asserts that 'ϵ_n* diverges from ϵ_ϕ' and approximates the optimal negative direction n*(p,t). This is not demonstrated. A DNS chain follows a different, negatively guided trajectory, so its noise at time t-K_t need not be the best negative direction for the main chain at time t. To support the central 'adaptive' claim, the authors should run a fixed-negative control: compute one DNS chain at t=T, normalize its noise, and reuse the same noise at every step. If the gains persist without per-step recomputation, the adaptivity is not the driver; if they vanish, the claim is supported. As written, the benefit attributed to adaptivity is not isolated.
  2. No error bars, confidence intervals, or significance tests are reported for any of the quantitative metrics or win rates. The differences between methods are often small (e.g., CLIP scores of 32.97 vs. 33.62 in Table 1), and without measures of variance it is impossible to judge whether these differences are reliable. The win rates are point estimates with no indication of the number of images, seeds, or prompts. The authors should report standard errors or confidence intervals over multiple seeds or prompt subsets, and ideally a paired significance test for the main ANSWER versus CFG and ANSWER versus DNP comparisons.
  3. The text states that 'SDXL+ANSWER outperforms SDXL (CFG) and SDXL+DNP across all metrics' for the ImageNet experiment and earlier claims similar broad superiority. This is contradicted by the Attend&Excite row in Table 1, where the IS value for ANSWER (11.15) is lower than that of DNP (11.65), as the preceding paragraph itself acknowledges. The summary sentence must be corrected to reflect the actual exception, and the discrepancy should be explained. A factual inconsistency in the main quantitative claim undermines confidence in the other reported results.
  4. The AMT human evaluation is described only by aggregate percentages in Figure 7. The paper does not report the number of participants, the number of pairwise comparisons, the number of prompts used, or whether the reported rates (ANSWER 46–61%, DNP 21–32%, CFG 8–20%) are conditioned on excluding 'No Clear Winner' responses. Without this information, the abstract's claim that ANSWER is 'preferred by humans 2x more over the other methods' cannot be verified. Please provide the full experimental protocol and, if possible, a significance test or confidence interval for the preference rates.
minor comments (6)
  1. The sentence 'For all datasets, SDXL+ANSWER outperforms SDXL (CFG) and SDXL+DNP across all metrics' should be qualified to exclude the IS metric on Attend&Excite, as noted in the major comments.
  2. The sampling update 'Sample(z, ϵ^t, t)' is not defined; the authors should specify whether they use DDIM, DDPM, or another sampler, since the DNS chain and the main chain must use the same sampler for the claimed equivalence to hold.
  3. The scheduler constants in Schdlr(K,t) = K·((2t-T)/(2t-T+20))·((T+20)/T) are introduced without justification or sensitivity analysis; please state how these constants were chosen and discuss their effect on the results.
  4. The claim that ANSWER has O(KT) complexity should be stated more precisely: each DNS step requires two model evaluations (ϵ_p and ϵ_ϕ), and the main NP step also requires two evaluations, so the total number of model evaluations is larger than the CFG baseline by a factor that depends on K and the scheduler. Reporting wall-clock time in Figure 5 would make the trade-off concrete.
  5. Reference [6] is cited as 'Improving image synthesis with diffusion-negative sampling, 2024' without a venue; please provide the published reference or arXiv identifier if available.
  6. The figure uses color to distinguish methods; adding textual labels or a colorblind-safe palette would improve accessibility, since the percentages are central to the human-preference claim.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: ANSWER's empirical claims are benchmark-tested, and the only algebraic reduction (K_t=1 to rescaled CFG) is explicitly disclosed.

full rationale

I examined the derivation chain from Eq. (8)'s definition of the optimal negative prompt, through the DNS equation (Eq. 9) imported from the authors' prior work, to the CNP construction (Eq. 12) and the final ANSWER algorithm. The only exact reduction in the paper is the disclosed K_t = 1 case, in which the composition of DNS and NP collapses to CFG with rescaled guidance; this is stated explicitly in Section 3.3 and is used as motivation for K_t > 1, not as a hidden source of the reported improvements. For K_t > 1, the negative noise is obtained by running the DNS chain from the current latent and then re-normalizing; nothing in these equations forces the downstream benchmark scores, which are measured on external datasets (ImageNet, Attend&Excite, Pick-a-Pic, DrawBench, PartiPrompts) and by human preference. The reliance on the authors' earlier DNP/DNS work for the optimality of DNS is a self-citation, but the cited equation is a parameter-free component with its own external evaluation, and the paper's novel contribution—adaptive, captioner-free reuse of DNS noise in the NP chain—is not equivalent to that component by construction. The lack of a fixed-negative-noise control is an experimental completeness issue, not a circularity, so no step reduces to its own input.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

All free parameters are settings of the sampling procedure (K, s_n, scheduler constants, switching point); none are fitted to the reported benchmark numbers, which reduces circularity but the theory does not predict them. The per-step 'negative noise' is a method variable, not a newly postulated physical entity. The central optimality claims rest on axioms inherited from the authors' own prior DNP/DNS work and one new ad-hoc assumption about truncated DNS chains.

free parameters (4)
  • K (number of DNS steps per NP iteration) = K=5 (empirically chosen)
    Section 5 ablations show K=15 blurs images and K=0 is plain CFG; the authors pick K=5 once and use it for all experiments. This is a hand-tuned hyperparameter controlling the strength of the correction.
  • Negative guidance scale s_n in DNS chain = not stated (presumably 1)
    Algorithm 1 uses s_n in the DNS update but the main text never lists the value used for s_n in any experiment; it is a free parameter.
  • Scheduler shape constants (20, (T+20)/T) = hand-chosen
    The scheduler Schdlr(K,t) is introduced without derivation or ablation of its exact form; the constants 20 and (T+20)/T are chosen by hand.
  • Switch threshold t > T/2 = T/2
    ANSWER is applied only for the first half of the diffusion chain; this cutoff is justified by a qualitative observation (Figure 4) and latent-editing citations, not by measurement.
assumptions (6)
  • standard math Energy-based interpretation: denoising network predicts negative score of log p(z_t | prompt)
    Section 2.1, used throughout to justify CFG and negative prompting.
  • domain assumption The optimal negative prompt maximizes the steepness of the odds ratio o(z,p,n) (Eq. 8)
    Adopted from the authors' prior DNP work [6]; underlies the DNS equation.
  • domain assumption The DNS update (Eq. 9) generates the optimal negative image
    Section 3.1; the paper cites [6] for details and does not re-derive DNS here.
  • ad hoc to paper A K-step truncated DNS chain from the current latent z_t produces a noise estimate that serves as the optimal negative at step t (Eq. 12 intuition)
    Section 3.3; the central heuristic of ANSWER, asserted without proof.
  • standard math Bayes decision rule for positive vs negative prompts (o >= 1 chooses positive)
    Section 2.2; standard decision theory used to justify maximizing odds ratio.
  • domain assumption Later diffusion steps only allow cosmetic edits, so ANSWER can stop at T/2
    Section 3.3, supported by citations [5,23] and the qualitative Figure 4; not measured in this paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Diffusion Models with Adaptive Negative Sampling Without External Resources." pith.science (2026). https://pith.science/paper/DLFNRTOF

@misc{pith2026250802973,
  author       = {Pith},
  title        = {Pith review of: Diffusion Models with Adaptive Negative Sampling Without External Resources},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DLFNRTOF}},
  note         = {Machine review of arXiv:2508.02973}
}
read the original abstract

Diffusion models (DMs) have demonstrated an unparalleled ability to create diverse and high-fidelity images from text prompts. However, they are also well-known to vary substantially regarding both prompt adherence and quality. Negative prompting was introduced to improve prompt compliance by specifying what an image must not contain. Previous works have shown the existence of an ideal negative prompt that can maximize the odds of the positive prompt. In this work, we explore relations between negative prompting and classifier-free guidance (CFG) to develop a sampling procedure, {\it Adaptive Negative Sampling Without External Resources} (ANSWER), that accounts for both positive and negative conditions from a single prompt. This leverages the internal understanding of negation by the diffusion model to increase the odds of generating images faithful to the prompt. ANSWER is a training-free technique, applicable to any model that supports CFG, and allows for negative grounding of image concepts without an explicit negative prompts, which are lossy and incomplete. Experiments show that adding ANSWER to existing DMs outperforms the baselines on multiple benchmarks and is preferred by humans 2x more over the other methods.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

39 extracted references · 28 canonical work pages

  1. [6]

    Improving image syn- thesis with diffusion-negative sampling, 2024

    Alakh Desai and Nuno Vasconcelos. Improving image syn- thesis with diffusion-negative sampling, 2024. 2, 4, 5, 8

    work page 2024

  2. [2]

    Segmentation-free guidance for text-to-image dif- fusion models, 2024

    Kambiz Azarian, Debasmit Das, Qiqi Hou, and Fatih Porikli. Segmentation-free guidance for text-to-image dif- fusion models, 2024. 8

    work page 2024

  3. [1]

    Negative prompt: Stable diffusion webui.https://github.com/AUTOMATIC1111/ stable - diffusion - webui / wiki / Negative - prompt

    AUTOMATIC1111. Negative prompt: Stable diffusion webui.https://github.com/AUTOMATIC1111/ stable - diffusion - webui / wiki / Negative - prompt. 3

    work page

  4. [3]

    Retrieval-augmented diffusion models

    Andreas Blattmann, Robin Rombach, Kaan Oktay, Jonas M¨uller, and Bj ¨orn Ommer. Retrieval-augmented diffusion models. InAdvances in Neural Information Processing Sys- tems, 2022. 8

    work page 2022

  5. [4]

    Coyo-700m: Image-text pair dataset.https : / / github

    Minwoo Byeon, Beomhee Park, Haecheon Kim, Sungjun Lee, Woonhyuk Baek, and Saehoon Kim. Coyo-700m: Image-text pair dataset.https : / / github . com / kakaobrain/coyo-dataset, 2022. 8

    work page 2022

  6. [5]

    Attend-and-excite: Attention-based se- mantic guidance for text-to-image diffusion models, 2023

    Hila Chefer, Yuval Alaluf, Yael Vinker, Lior Wolf, and Daniel Cohen-Or. Attend-and-excite: Attention-based se- mantic guidance for text-to-image diffusion models, 2023. 2, 6, 8

    work page 2023

  7. [7]

    Compositional visual generation with energy based models

    Yilun Du, Shuang Li, and Igor Mordatch. Compositional visual generation with energy based models. InAdvances in Neural Information Processing Systems, pages 6637–6647. Curran Associates, Inc., 2020. 5

    work page 2020

  8. [8]

    Training-free structured diffusion guidance for compositional text-to-image synthesis, 2023

    Weixi Feng, Xuehai He, Tsu-Jui Fu, Varun Jampani, Arjun Akula, Pradyumna Narayana, Sugato Basu, Xin Eric Wang, and William Yang Wang. Training-free structured diffusion guidance for compositional text-to-image synthesis, 2023. 8

    2023

Show all 39 references
  1. [9]

    Expressive text-to-image generation with rich text

    Songwei Ge, Taesung Park, Jun-Yan Zhu, and Jia-Bin Huang. Expressive text-to-image generation with rich text. InIEEE International Conference on Computer Vision (ICCV), 2023. 8

    2023

  2. [10]

    Clipscore: A reference-free evaluation met- ric for image captioning, 2022

    Jack Hessel, Ari Holtzman, Maxwell Forbes, Ronan Le Bras, and Yejin Choi. Clipscore: A reference-free evaluation met- ric for image captioning, 2022. 6

    2022

  3. [11]

    Gans trained by a two time-scale update rule converge to a local nash equilib- rium.Advances in neural information processing systems, 30, 2017

    Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, and Sepp Hochreiter. Gans trained by a two time-scale update rule converge to a local nash equilib- rium.Advances in neural information processing systems, 30, 2017. 6

    2017

  4. [12]

    Classifier-free diffusion guidance.arXiv preprint arXiv:2207.12598, 2022

    Jonathan Ho and Tim Salimans. Classifier-free diffusion guidance.arXiv preprint arXiv:2207.12598, 2022. 2, 3, 8

    2022 arXiv

  5. [13]

    Denoising dif- fusion probabilistic models.Advances in neural information processing systems, 33:6840–6851, 2020

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising dif- fusion probabilistic models.Advances in neural information processing systems, 33:6840–6851, 2020. 3

    2020

  6. [14]

    Elucidating the design space of diffusion-based generative models, 2022

    Tero Karras, Miika Aittala, Timo Aila, and Samuli Laine. Elucidating the design space of diffusion-based generative models, 2022. 3

    2022

  7. [15]

    Pick-a-pic: An open dataset of user preferences for text-to-image generation

    Yuval Kirstain, Adam Polyak, Uriel Singer, Shahbuland Ma- tiana, Joe Penna, and Omer Levy. Pick-a-pic: An open dataset of user preferences for text-to-image generation

  8. [16]

    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. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1931–1941, 2023. 8

    1931

  9. [17]

    Gligen: Open-set grounded text-to-image generation

    Yuheng Li, Haotian Liu, Qingyang Wu, Fangzhou Mu, Jian- wei Yang, Jianfeng Gao, Chunyuan Li, and Yong Jae Lee. Gligen: Open-set grounded text-to-image generation. InPro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 22511–22521, 2023. 1, 8

    2023

  10. [18]

    Compositional visual generation with composable diffusion models

    Nan Liu, Shuang Li, Yilun Du, Antonio Torralba, and Joshua B Tenenbaum. Compositional visual generation with composable diffusion models. InEuropean Conference on Computer Vision, pages 423–439. Springer, 2022. 2, 3, 8

    2022

  11. [19]

    Grounded text-to-image synthesis with attention refocusing.arXiv preprint arXiv:2306.05427, 2023

    Quynh Phung, Songwei Ge, and Jia-Bin Huang. Grounded text-to-image synthesis with attention refocusing.arXiv preprint arXiv:2306.05427, 2023. 1, 8

    2023 arXiv

  12. [20]

    Sdxl: Improving latent diffusion models for high-resolution image synthesis, 2023

    Dustin Podell, Zion English, Kyle Lacey, Andreas Blattmann, Tim Dockhorn, Jonas M ¨uller, Joe Penna, and Robin Rombach. Sdxl: Improving latent diffusion models for high-resolution image synthesis, 2023. 1, 8

    2023

  13. [21]

    Zero-shot text-to-image generation

    Aditya Ramesh, Mikhail Pavlov, Gabriel Goh, Scott Gray, Chelsea V oss, Alec Radford, Mark Chen, and Ilya Sutskever. Zero-shot text-to-image generation. InInternational Confer- ence on Machine Learning, pages 8821–8831. PMLR, 2021

    2021

  14. [22]

    Hierarchical text-conditional image gener- ation with clip latents.arXiv preprint arXiv:2204.06125, 1 (2):3, 2022

    Aditya Ramesh, Prafulla Dhariwal, Alex Nichol, Casey Chu, and Mark Chen. Hierarchical text-conditional image gener- ation with clip latents.arXiv preprint arXiv:2204.06125, 1 (2):3, 2022. 1, 8

    2022 arXiv

  15. [23]

    Linguistic bind- ing in diffusion models: Enhancing attribute correspondence through attention map alignment, 2023

    Royi Rassin, Eran Hirsch, Daniel Glickman, Shauli Rav- fogel, Yoav Goldberg, and Gal Chechik. Linguistic bind- ing in diffusion models: Enhancing attribute correspondence through attention map alignment, 2023. 2, 6, 8

    2023

  16. [24]

    High-resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj ¨orn Ommer. High-resolution image synthesis with latent diffusion models. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022. 8

    2022

  17. [25]

    Dreambooth: Fine tuning text-to-image diffusion models for subject-driven generation

    Nataniel Ruiz, Yuanzhen Li, Varun Jampani, Yael Pritch, Michael Rubinstein, and Kfir Aberman. Dreambooth: Fine tuning text-to-image diffusion models for subject-driven generation. 2022. 8

    2022

  18. [26]

    Berg, and Li Fei-Fei

    Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, San- jeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, Alexander C. Berg, and Li Fei-Fei. ImageNet Large Scale Visual Recognition Chal- lenge.International Journal of Computer Vision (...

    2015

  19. [27]

    Photorealistic text-to-image diffusion models with deep language understanding.Advances in Neural Information Processing Systems, 35:36479–36494, 2022

    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

  20. [28]

    Improved techniques for training gans, 2016

    Tim Salimans, Ian Goodfellow, Wojciech Zaremba, Vicki Cheung, Alec Radford, and Xi Chen. Improved techniques for training gans, 2016. 6

    2016

  21. [29]

    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 Worts- man, Patrick Schramowski, Srivatsa Kundurthy, Katherine Crowson, Ludwig Schmidt, Robert Kaczmarczyk, and Jenia Jitsev. Lai...

    2022

  22. [30]

    Denoising diffusion implicit models.arXiv preprint arXiv:2010.02502, 2020

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denoising diffusion implicit models.arXiv preprint arXiv:2010.02502, 2020. 3

    2010 arXiv

  23. [31]

    Generative modeling by esti- mating gradients of the data distribution.Advances in neural information processing systems, 32, 2019

    Yang Song and Stefano Ermon. Generative modeling by esti- mating gradients of the data distribution.Advances in neural information processing systems, 32, 2019. 3

    2019

  24. [32]

    Adapting diffusion models for improved prompt compliance and controllable image synthesis, 2024

    Deepak Sridhar, Abhishek Peri, Rohith Rachala, and Nuno Vasconcelos. Adapting diffusion models for improved prompt compliance and controllable image synthesis, 2024. 8

    2024

  25. [33]

    Sketch-guided text-to-image diffusion models, 2022

    Andrey V oynov, Kfir Aberman, and Daniel Cohen-Or. Sketch-guided text-to-image diffusion models, 2022. 1

    2022

  26. [34]

    Bayesian learning via stochas- tic gradient langevin dynamics

    Max Welling and Yee W Teh. Bayesian learning via stochas- tic gradient langevin dynamics. InProceedings of the 28th international conference on machine learning (ICML-11), pages 681–688, 2011. 3

    2011

  27. [35]

    Human preference score v2: A solid benchmark for evaluating human preferences of text-to-image synthesis, 2023

    Xiaoshi Wu, Yiming Hao, Keqiang Sun, Yixiong Chen, Feng Zhu, Rui Zhao, and Hongsheng Li. Human preference score v2: A solid benchmark for evaluating human preferences of text-to-image synthesis, 2023. 6

    2023

  28. [36]

    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. InProceedings of the 37th International Confer- ence on Neural Information Processing Systems, R...

    2024

  29. [37]

    Scaling autoregressive models for content-rich text-to-image generation, 2022

    Jiahui Yu, Yuanzhong Xu, Jing Yu Koh, Thang Luong, Gun- jan Baid, Zirui Wang, Vijay Vasudevan, Alexander Ku, Yin- fei Yang, Burcu Karagol Ayan, Ben Hutchinson, Wei Han, Zarana Parekh, Xin Li, Han Zhang, Jason Baldridge, and Yonghui Wu. Scaling autoregressive models for content...

    2022

  30. [38]

    Adding conditional control to text-to-image diffusion models, 2023

    Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. Adding conditional control to text-to-image diffusion models, 2023. 1, 8

    2023

  31. [39]

    Layoutdiffusion: Controllable diffu- sion model for layout-to-image generation

    Guangcong Zheng, Xianpan Zhou, Xuewei Li, Zhongang Qi, Ying Shan, and Xi Li. Layoutdiffusion: Controllable diffu- sion model for layout-to-image generation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, pages 22490–22499, 2023. 1, 8

    2023

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.