Pith. sign in

REVIEW 4 major objections 6 minor 53 references

Stable Score Distillation

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 Stable Score Distillation adds a null-text branch to the editing score, producing text-guided 2D and 3D edits that preserve source structure and align more closely with the target prompt than DDS or CSD.

desk verdict The paper's main novelty is a constant offset with zero gradient, so the cross-trajectory term does nothing to the optimization and the stability claim is unsupported. read the letter →

arxiv 2507.09168 v1 pith:RUATH55L submitted 2025-07-12 cs.CV

Haiming Zhu , Yangyang Xu , Chenshu Xu , Tingrui Shen , Wenxi Liu , Yong Du , Jun Yu , Shengfeng He This is my paper
classification cs.CV
keywords text-guidedimageediting3Dscenescoredistillationclassifier-freeguidancenull-textbranchdiffusionmodelsNeRFGaussiansplatting
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

Text-guided editing of images and 3D scenes usually relies on score distillation losses, but existing formulations like $\Delta$ Denoising Score (DDS) and Classifier Score Distillation (CSD) can blur unedited regions or deform structure. This paper introduces Stable Score Distillation (SSD), a loss that adds a null-text branch to the editing score. The loss combines a cross-prompt Classifier-Free Guidance (CFG) term with a cross-trajectory null-text term and a prompt-enhancement term. The authors report that SSD preserves the source structure and aligns edits with the target prompt more reliably than DDS and CSD on 2D and 3D editing benchmarks, while converging in fewer iterations.

What carries the argument

The load-bearing object is the SSD score of Eq. 5, $\mathcal{L}_{\mathrm{ssd}} = \epsilon_\phi(z_t, \hat{y}) + s(\epsilon_\phi(z_t, y) - \epsilon_\phi(z_t, \hat{y})) - \epsilon_\phi(\hat{z}_t, \emptyset)$. The first two terms form the cross-prompt CFG direction from source prompt $\hat{y}$ to target prompt $y$ on the current latent $z_t$; the third term, a null-text prediction on the source latent $\hat{z}_t$, is the paper's main novelty and is decomposed into the cross-trajectory term. The full loss adds a prompt-enhancement branch $w_e(\epsilon_\phi(z_t, y) - \epsilon_\phi(z_t, \emptyset))$ and an ID regularization term that uses a noisy source latent to avoid exploding gradients in 3D Gaussian splatting. The cross-trajectory term is said to measure the distance between the transitions of two latents, which stabilizes optimization by keeping the structure smooth.

What would settle it

Sweep the weight of the null-text term $\epsilon_\phi(\hat{z}_t, \emptyset)$ in Eq. 5 while holding all other terms fixed, across a set of 2D and 3D editing tasks, and measure structure-preservation (DINO, LPIPS) and prompt-alignment (CLIP) metrics. If no consistent improvement or monotonic trend appears relative to weight zero, the term is not acting as a structure regularizer. Alternatively, set the term to a constant zero; if the edited results are unchanged, the paper's central novelty has no causal effect.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that a score distillation loss built from three terms—a cross-prompt CFG direction, a cross-trajectory null-text term, and a prompt-enhancement term—produces text-guided 2D and 3D edits that are more stable and more aligned with the target prompt than those from DDS and CSD. The decisive element is the cross-trajectory term $\epsilon_\phi(\hat{z}_t, \emptyset)$, which the paper interprets as a structure-preserving regularizer that keeps the optimization trajectory close to the source content. The resulting method also converges in about 3,000 iterations for NeRF and 1,500 for 3D Gaussian splatting, with no auxiliary source branch.

Load-bearing premise

The load-bearing premise is that subtracting the diffusion model's null-text prediction on the source latent is a faithful, unbiased 'no-edit' reference; if that term instead injects an uncontrolled bias, the method collapses to Classifier Score Distillation plus an arbitrary offset.

Editorial extensions

If this is right

  • Pipelines based on DDS can drop the source branch, since the null-text term supplies the structure-preserving signal.
  • The method works directly on NeRF and 3D Gaussian splatting scenes, converging in about 3,000 and 1,500 iterations respectively.
  • Style edits that DDS-based methods often fail to apply become feasible with the prompt-enhancement branch.
  • The framework connects to InstructPix2Pix, showing that its one-step reverse sampling is, in effect, a simple version of the SSD loss with the cross-trajectory and cross-prompt terms.
  • The streamlined design requires no LoRA or fine-tuning and integrates with Stable Diffusion and InstructPix2Pix.

Reading between the lines

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

  • If the null-text term truly acts as a no-edit reference, the same regularization principle could be transferred to other SDS-based tasks such as text-to-3D generation, where source-structure preservation is less defined.
  • The paper's reading of InstructPix2Pix suggests that other one-step editing models may be decomposable into cross-prompt and cross-trajectory components, which could guide future loss designs.
  • Extending the paper's ablation to a full weight sweep would reveal whether the cross-trajectory term behaves monotonically as a structure regularizer, since only a few hand-selected weights are currently shown.
  • The method may struggle when the source image is far from the diffusion model's training distribution, because the null-text prediction then becomes unreliable as a 'no-edit' reference.
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 Stable Score Distillation (SSD), a score distillation loss for text-guided editing of both 2D images and 3D scenes (NeRF/3DGS). The loss replaces the source branch of DDS with a CFG-based cross-prompt term, adds a null-text term evaluated on the source latent (called a cross-trajectory term), and supplements these with a prompt-enhancement term and an ID regularization. The authors claim that this design improves stability, prompt alignment, source-content preservation, and convergence speed, and they report experiments on 3D scene editing (IN2N, LLFF, Mip-NeRF360) and 2D editing (PIE-Bench) against DDS, CSD, and other baselines. The paper includes qualitative comparisons, quantitative tables, a user study, and a discussion connecting the method to InstructPix2Pix.

Significance. If the central claims are correct, SSD would provide a simpler and more stable alternative to DDS/CSD for text-guided editing, with the additional conceptual contribution of linking score distillation to InstructPix2Pix. The paper is clearly written in structure, releases code, and identifies a concrete design space. However, the load-bearing theoretical justification for the null-text term is questionable from a gradient perspective, and the quantitative evidence is not fully consistent with the stated claims, particularly regarding structure preservation in Table 2. The absence of quantitative ablations and error bars further limits the strength of the empirical support. The work is potentially useful, but the manuscript in its current form does not convincingly establish the claimed advantages.

major comments (4)
  1. [Sec. 4.1, Eqs. (5)-(6)] The central novelty is the cross-trajectory term -epsilon_phi(z_hat_t, empty) in Eq. (5), but in the SDS update (Eq. (1)) the source latent z_hat_t is fixed with respect to the optimized parameters theta. Therefore this term contributes a theta-independent offset to the loss and zero to the gradient, so it cannot by itself enforce source-structure preservation or stabilize the optimization. In the decomposition of Eq. (6), the only theta-dependent part of the cross-trajectory term is w_t * epsilon_phi(z_t, y_hat), which merely reweights the source-prompt score rather than introducing a new constraint. The paper should provide a derivation showing how this term affects the optimization trajectory (for example through the fixed point) or revise the claim that it regularizes structure.
  2. [Sec. 5.2, Table 2] Table 2 directly contradicts the central claim of source-structure preservation. Full SSD has DINO distance 28.13 x 10^-3, LPIPS 82.43 x 10^-3, and MSE 86.64 x 10^-4, versus 7.15, 33.14, and 25.29 for DDS+CDS, respectively, i.e., roughly 4x, 2.5x, and 3.4x worse. The text in Sec. 5.2 describes this as 'a slightly higher structure distance,' which understates the magnitude. If the method's main selling point is preserving source structure, this quantitative result needs to be reconciled, for example by reporting structure metrics on the edited regions only or by demonstrating that the larger distance comes from intended edits rather than unintended changes.
  3. [Sec. 5.3, Fig. 5] The ablations for the cross-trajectory and prompt-enhancement terms are qualitative only in the main text, with hand-selected weights (w_t = 0, 1.0, 1.5, 2.0 and w_e = 0, 1.5, 5.5, 7.5) shown in Fig. 5. No quantitative metrics are reported, and the text refers to 'the Supp.' for quantitative evaluation, but no supplementary material is included in the arXiv submission. Without a quantitative ablation that isolates the contribution of the null-text term, and without error bars or significance tests, the claim that this term provides stability and structure preservation is not empirically supported.
  4. [Sec. 4.1, Eq. (5) vs Fig. 5] The loss in Eq. (5) fixes the coefficient of the null-text term to 1, which corresponds to w_t = 1 in the decomposition of Eq. (6). However, Fig. 5 sweeps w_t over values 0, 1.5, and 2.0, so the empirical study is not evaluating the proposed loss in Eq. (5) but a family of losses with a free parameter. The relationship between the theoretical loss and the implemented loss should be clarified, and the final reported results should state which weight values were used.
minor comments (6)
  1. [Abstract] The abstract contains grammatical issues: 'achieves' should be 'achieve' and 'introduces a constant term null-text branch' is awkward; please rephrase for clarity.
  2. [Sec. 4.1, Eq. (4)] Equation (4) uses 'Grad' as a variable without defining whether it is a loss or a gradient, and 'The scale factor s is equal to control weight' is vague; please specify the role of s and its allowed range.
  3. [Sec. 4.3, Eq. (8)] The notation in Eq. (8) is unclear: x_t and z_hat_t are not explicitly defined; please state that x_t is the noised rendered latent and z_hat_t is the noised source latent, and clarify how the time indices are aligned.
  4. [Sec. 4.4] There are typos in Sec. 4.4 ('ours edit gard provide new angle') and inconsistent naming of InstructPix2Pix (Instructpix2pix, InstructP2P); please unify the terminology and proofread.
  5. [Sec. 5.1, Table 1] The user study in Table 1 reports a single percentage without details on the number of questions, the rating scale, or statistical significance; please provide confidence intervals or a significance test.
  6. [Fig. 5] The horizontal axis of Fig. 5 is not labeled; it is unclear what quantity is being varied (e.g., iteration number or weight value). Please add axis labels and a legend.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity found: SSD combines known score-distillation terms into a new heuristic loss, and its evaluations are external rather than derived from the loss by construction.

full rationale

The paper's central object, Eq.5, is proposed as a new loss, not derived from a theorem. Eq.4 explicitly restates Classifier-Free Guidance (CFG) with the source prompt as the base condition and a guidance scale s; using a published, external formula as a component is not circular. The null-text branch epsilon_phi(z_hat_t,null) is introduced as an ansatz to stabilize optimization; the paper's claim that it preserves structure is supported only by qualitative ablations (Fig.5) and is contradicted by the higher structure-distance numbers in Table 2 for the full model. That is an evidence or correctness problem, not a circularity: the loss is not defined in terms of the claimed outcome, and no fitted parameter is later relabeled as a prediction. Eq.6 is a reparameterization of Eq.5 with explicit weights, not a separate derived result. The references to DDS, CSD, NFSD, and InstructPix2Pix are external prior works, not self-citations. No uniqueness theorem or prior result by the present authors is invoked to force the choice of Eq.5. Consequently, no step in the paper reduces to its own inputs by construction, and the appropriate finding is no significant circularity (score 0).

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

The method depends on four tuned weights and several empirical assumptions about score differences acting as structure regularizers. No new physical or ontological entities are introduced; the 'cross-trajectory term' is a named loss component, not an independently evidenced entity.

free parameters (4)
  • cross-prompt weight w_p (s) = 7.5 (fixed in Fig.5 sweep)
    Controls the CFG interpolation strength in Eq.4; chosen by visual inspection, not by held-out validation.
  • cross-trajectory weight w_t = range 0-2, final value not stated
    Controls the null-text regularization term in Eq.6; swept in Fig.5 without specified selection criterion.
  • prompt enhancement weight w_e = range 0-7.5, final value not stated
    Controls the target alignment branch in Eq.7; swept in Fig.5, values differ for style vs content edits, no automatic schedule.
  • ID regularization strength w(t) = decreasing function of t, not specified
    Used in Eq.8; the paper says it is not necessary to be well-designed, yet its weight or schedule is not given and it changes results in Fig.8.
assumptions (4)
  • standard math Diffusion model noise prediction equals the score function, and CFG provides a direction toward the conditional density.
    Invoked to justify using CFG differences as editing gradients in Sec.2.1 and Eq.4.
  • ad hoc to paper Subtracting the null-text score on the source latent, epsilon_phi(z_hat, null), preserves source structure without introducing bias.
    This is the core stability assumption in Eq.5; it is asserted in Sec.4.1 and not derived or formally tested.
  • ad hoc to paper The cross-trajectory distance (epsilon_phi(z_t, y_hat) - epsilon_phi(z_hat, null)) measures structural deviation between latents.
    Used to claim structure preservation in Sec.4.1; no metric or theoretical link to pixel-level structure is provided.
  • domain assumption A linear interpolation between source and target prompt scores yields a gradual and smooth editing direction.
    Eq.4 assumes CFG-style interpolation is monotonic and stable for editing; the paper provides only qualitative illustrations.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Stable Score Distillation." pith.science (2026). https://pith.science/paper/RUATH55L

@misc{pith2026250709168,
  author       = {Pith},
  title        = {Pith review of: Stable Score Distillation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RUATH55L}},
  note         = {Machine review of arXiv:2507.09168}
}
read the original abstract

Text-guided image and 3D editing have advanced with diffusion-based models, yet methods like Delta Denoising Score often struggle with stability, spatial control, and editing strength. These limitations stem from reliance on complex auxiliary structures, which introduce conflicting optimization signals and restrict precise, localized edits. We introduce Stable Score Distillation (SSD), a streamlined framework that enhances stability and alignment in the editing process by anchoring a single classifier to the source prompt. Specifically, SSD utilizes Classifier-Free Guidance (CFG) equation to achieves cross-prompt alignment, and introduces a constant term null-text branch to stabilize the optimization process. This approach preserves the original content's structure and ensures that editing trajectories are closely aligned with the source prompt, enabling smooth, prompt-specific modifications while maintaining coherence in surrounding regions. Additionally, SSD incorporates a prompt enhancement branch to boost editing strength, particularly for style transformations. Our method achieves state-of-the-art results in 2D and 3D editing tasks, including NeRF and text-driven style edits, with faster convergence and reduced complexity, providing a robust and efficient solution for text-guided editing.

Figures

Figures reproduced from arXiv: 2507.09168 by the authors.

Figure 1
Figure 1. We propose Stable Score Distillation (SSD), a method that improves text-guided editing by preserving original content structure [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Illustration of three distillation-based approaches. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. The optimization process of DDS and our SSD. SSD [PITH_FULL_IMAGE:figures/full_fig_p002_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: The overview of SSD. Given the parameter 3D-model or image, SSD provides effective editing gradient to guide the optimization [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: The effect of increasing the strength of the prompt en [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: Qualitative comparisons with related works. SDS demonstrates outstanding performance in effectively preserve source structure in [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 8
Figure 8. Figure 8: Effect of source latent regularization. In most experi [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 7
Figure 7. Figure 7: Comparison of different editing methods on various [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

53 extracted references · 37 canonical work pages

  1. [1]

    Mip-nerf 360: Unbounded anti-aliased neural radiance fields

    Jonathan T Barron, Ben Mildenhall, Dor Verbin, Pratul P Srinivasan, and Peter Hedman. Mip-nerf 360: Unbounded anti-aliased neural radiance fields. InCVPR, pages 5470– 5479, 2022. 6

    work page 2022

  2. [2]

    In- structpix2pix: Learning to follow image editing instructions

    Tim Brooks, Aleksander Holynski, and Alexei A Efros. In- structpix2pix: Learning to follow image editing instructions. InCVPR, pages 18392–18402, 2023. 3, 6

    work page 2023

  3. [3]

    Masactrl: Tuning-free mutual self-attention control for consistent image synthesis and edit- ing

    Mingdeng Cao, Xintao Wang, Zhongang Qi, Ying Shan, Xi- aohu Qie, and Yinqiang Zheng. Masactrl: Tuning-free mutual self-attention control for consistent image synthesis and edit- ing. InICCV, pages 22560–22570, 2023. 6

    work page 2023

  4. [4]

    Emerg- ing properties in self-supervised vision transformers

    Mathilde Caron, Hugo Touvron, Ishan Misra, Herv´e J´egou, Julien Mairal, Piotr Bojanowski, and Armand Joulin. Emerg- ing properties in self-supervised vision transformers. InICCV,

  5. [5]

    Dge: Direct gaussian 3d editing by consistent multi-view editing.arXiv preprint arXiv:2404.18929, 2024

    Minghao Chen, Iro Laina, and Andrea Vedaldi. Dge: Direct gaussian 3d editing by consistent multi-view editing.arXiv preprint arXiv:2404.18929, 2024. 6

    arXiv 2024

  6. [6]

    Gaussianeditor: Swift and controllable 3d editing with gaussian splatting

    Yiwen Chen, Zilong Chen, Chi Zhang, Feng Wang, Xiaofeng Yang, Yikai Wang, Zhongang Cai, Lei Yang, Huaping Liu, and Guosheng Lin. Gaussianeditor: Swift and controllable 3d editing with gaussian splatting. InCVPR, pages 21476–21485,

    work page

  7. [7]

    Turboedit: Text-based image editing using few-step diffusion models

    Gilad Deutch, Rinon Gal, Daniel Garibi, Or Patashnik, and Daniel Cohen-Or. Turboedit: Text-based image editing using few-step diffusion models. pages 1–12, 2024. 5, 8

    work page 2024

  8. [8]

    Diffusion models beat gans on image synthesis.NeurIPS, 34:8780–8794, 2021

    Prafulla Dhariwal and Alexander Nichol. Diffusion models beat gans on image synthesis.NeurIPS, 34:8780–8794, 2021. 3

    work page 2021

Show all 53 references
  1. [9]

    Instruct-nerf2nerf: Editing 3d scenes with instructions

    Ayaan Haque, Matthew Tancik, Alexei Efros, Aleksander Holynski, and Angjoo Kanazawa. Instruct-nerf2nerf: Editing 3d scenes with instructions. InICCV, 2023. 3, 6

    2023

  2. [10]

    Delta denoising score

    Amir Hertz, Kfir Aberman, and Daniel Cohen-Or. Delta denoising score. InICCV, pages 2328–2337, 2023. 1, 2, 3, 4, 6, 7

    2023

  3. [11]

    Prompt-to-prompt image editing with cross-attention control

    Amir Hertz, Ron Mokady, Jay Tenenbaum, Kfir Aberman, Yael Pritch, and Daniel Cohen-or. Prompt-to-prompt image editing with cross-attention control. InICLR, 2023. 6

    2023

  4. [12]

    Classifier-free diffusion guidance

    Jonathan Ho and Tim Salimans. Classifier-free diffusion guidance. InNeurIPS Workshop, 2021. 1, 3, 5

    2021

  5. [13]

    Denoising diffu- sion probabilistic models.NeurIPS, 33:6840–6851, 2020

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffu- sion probabilistic models.NeurIPS, 33:6840–6851, 2020. 1, 3

    2020

  6. [14]

    Lora: Low-rank adaptation of large language models

    Edward J Hu, yelong shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. Lora: Low-rank adaptation of large language models. InICLR,

  7. [15]

    Dreamtime: An improved optimization strategy for diffusion-guided 3d generation

    Yukun Huang, Jianan Wang, Yukai Shi, Boshi Tang, Xianbiao Qi, and Lei Zhang. Dreamtime: An improved optimization strategy for diffusion-guided 3d generation. InICLR, 2023. 3

    2023

  8. [16]

    An edit friendly ddpm noise space: Inversion and manipulations

    Inbar Huberman-Spiegelglas, Vladimir Kulikov, and Tomer Michaeli. An edit friendly ddpm noise space: Inversion and manipulations. InCVPR, pages 12469–12478, 2024. 3

    2024

  9. [17]

    Estimation of non- normalized statistical models by score matching.JMLR, 6(4),

    Aapo Hyv ¨arinen and Peter Dayan. Estimation of non- normalized statistical models by score matching.JMLR, 6(4),

  10. [18]

    Pnp inversion: Boosting diffusion-based editing with 3 lines of code

    Xuan Ju, Ailing Zeng, Yuxuan Bian, Shaoteng Liu, and Qiang Xu. Pnp inversion: Boosting diffusion-based editing with 3 lines of code. InICLR, 2024. 2, 6, 7

    2024

  11. [19]

    Noise-free score distillation

    Oren Katzir, Or Patashnik, Daniel Cohen-Or, and Dani Lischinski. Noise-free score distillation. InICLR, 2024. 1, 2, 4

    2024

  12. [20]

    3d gaussian splatting for real-time radiance field rendering.ACM TOG, 42(4):1–14, 2023

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimk¨uhler, and George Drettakis. 3d gaussian splatting for real-time radiance field rendering.ACM TOG, 42(4):1–14, 2023. 3, 6

    2023

  13. [21]

    Dreamcatalyst: Fast and high-quality 3d editing via controlling editability and identity preservation.arXiv preprint arXiv:2407.11394, 2024

    Jiwook Kim, Seonho Lee, Jaeyo Shin, Jiho Choi, and Hyun- jung Shim. Dreamcatalyst: Fast and high-quality 3d editing via controlling editability and identity preservation.arXiv preprint arXiv:2407.11394, 2024. 2, 3

    2024 arXiv

  14. [22]

    Dream-in-style: Text-to-3d generation using stylized score distillation.arXiv preprint arXiv:2406.18581, 2024

    Hubert Kompanowski and Binh-Son Hua. Dream-in-style: Text-to-3d generation using stylized score distillation.arXiv preprint arXiv:2406.18581, 2024. 2

    2024 arXiv

  15. [23]

    Posterior distilla- tion sampling

    Juil Koo, Chanho Park, and Minhyuk Sung. Posterior distilla- tion sampling. InCVPR, pages 13352–13361, 2024. 2, 3, 5, 6

    2024

  16. [24]

    Preserving identity with variational score for general-purpose 3d editing.arXiv preprint arXiv:2406.08953, 2024

    Duong H Le, Tuan Pham, Aniruddha Kembhavi, Stephan Mandt, Wei-Chiu Ma, and Jiasen Lu. Preserving identity with variational score for general-purpose 3d editing.arXiv preprint arXiv:2406.08953, 2024. 3

    2024 arXiv

  17. [25]

    Learning pseudo 3d guidance for view-consistent texturing with 2d diffusion

    Kehan Li, Yanbo Fan, Yang Wu, Zhongqian Sun, Wei Yang, Xiangyang Ji, Li Yuan, and Jie Chen. Learning pseudo 3d guidance for view-consistent texturing with 2d diffusion. In ECCV, pages 18–34, 2025. 1

    2025

  18. [26]

    Styledif- fusion: Prompt-embedding inversion for text-based editing

    Senmao Li, Joost Van De Weijer, Taihang Hu, Fahad Shahbaz Khan, Qibin Hou, Yaxing Wang, and Jian Yang. Styledif- fusion: Prompt-embedding inversion for text-based editing. arXiv preprint arXiv:2303.15649, 2023. 7, 8

    2023 arXiv

  19. [27]

    Latent consistency models: Synthesizing high- resolution images with few-step inference.arXiv preprint arXiv:2310.04378, 2023

    Simian Luo, Yiqin Tan, Longbo Huang, Jian Li, and Hang Zhao. Latent consistency models: Synthesizing high- resolution images with few-step inference.arXiv preprint arXiv:2310.04378, 2023. 8

    2023 arXiv

  20. [28]

    Latent-nerf for shape-guided generation of 3d shapes and textures

    Gal Metzer, Elad Richardson, Or Patashnik, Raja Giryes, and Daniel Cohen-Or. Latent-nerf for shape-guided generation of 3d shapes and textures. InCVPR, pages 12663–12673, 2023. 1

    2023

  21. [29]

    Local light field fusion: Practical view synthe- sis with prescriptive sampling guidelines.ACM TOG, 38(4): 1–14, 2019

    Ben Mildenhall, Pratul P Srinivasan, Rodrigo Ortiz-Cayon, Nima Khademi Kalantari, Ravi Ramamoorthi, Ren Ng, and Abhishek Kar. Local light field fusion: Practical view synthe- sis with prescriptive sampling guidelines.ACM TOG, 38(4): 1–14, 2019. 6

    2019

  22. [30]

    Nerf: Representing scenes as neural radiance fields for view synthe- sis

    Ben Mildenhall, Pratul P Srinivasan, Matthew Tancik, Jonathan T Barron, Ravi Ramamoorthi, and Ren Ng. Nerf: Representing scenes as neural radiance fields for view synthe- sis. InECCV, pages 405–421, 2020. 3

    2020

  23. [31]

    Null-text inversion for editing real images using guided diffusion models

    Ron Mokady, Amir Hertz, Kfir Aberman, Yael Pritch, and Daniel Cohen-Or. Null-text inversion for editing real images using guided diffusion models. InCVPR, pages 6038–6047,

  24. [32]

    Contrastive denoising score for text-guided latent diffu- sion image editing

    Hyelin Nam, Gihyun Kwon, Geon Yeong Park, and Jong Chul Ye. Contrastive denoising score for text-guided latent diffu- sion image editing. InCVPR, pages 9192–9201, 2024. 2, 7, 8 9

    2024

  25. [33]

    ED-NeRF: Efficient text-guided editing of 3d scene with latent space nerf

    JangHo Park, Gihyun Kwon, and Jong Chul Ye. ED-NeRF: Efficient text-guided editing of 3d scene with latent space nerf. InICLR, 2024. 2, 3

    2024

  26. [34]

    Barron, and Ben Mildenhall

    Ben Poole, Ajay Jain, Jonathan T. Barron, and Ben Mildenhall. Dreamfusion: Text-to-3d using 2d diffusion. InICLR, 2023. 1, 3, 4

    2023

  27. [35]

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

    2022 arXiv

  28. [36]

    Ul- trapixel: Advancing ultra high-resolution image synthesis to new peaks

    Jingjing Ren, Wenbo Li, Haoyu Chen, Renjing Pei, Bin Shao, Yong Guo, Long Peng, Fenglong Song, and Lei Zhu. Ul- trapixel: Advancing ultra high-resolution image synthesis to new peaks. InNeurIPS, pages 111131–111171, 2024. 3

    2024

  29. [37]

    Turbo2k: Towards ultra-efficient and high-quality 2k video synthesis.arXiv preprint arXiv:2504.14470, 2025

    Jingjing Ren, Wenbo Li, Zhongdao Wang, Haoze Sun, Bangzhen Liu, Haoyu Chen, Jiaqi Xu, Aoxue Li, Shifeng Zhang, Bin Shao, et al. Turbo2k: Towards ultra-efficient and high-quality 2k video synthesis.arXiv preprint arXiv:2504.14470, 2025. 3

    2025 arXiv

  30. [38]

    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. InCVPR, pages 10684– 10695, 2022. 1, 2, 3, 6

    2022

  31. [39]

    Photorealistic text-to-image diffusion models with deep lan- guage 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 lan- guage understanding. InNeurIPS, pages 36479–36494...

    2022

  32. [40]

    Adversarial diffusion distillation

    Axel Sauer, Dominik Lorenz, Andreas Blattmann, and Robin Rombach. Adversarial diffusion distillation. InECCV, pages 87–103, 2025. 8

    2025

  33. [41]

    Denoising diffusion implicit models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denoising diffusion implicit models. InICLR, 2021. 1, 3, 6

    2021

  34. [42]

    Score-based generative modeling through stochastic differential equations

    Yang Song, Jascha Sohl-Dickstein, Diederik P Kingma, Ab- hishek Kumar, Stefano Ermon, and Ben Poole. Score-based generative modeling through stochastic differential equations. InICLR, 2021. 3

    2021

  35. [43]

    Postedit: Posterior sam- pling for efficient zero-shot image editing.arXiv preprint arXiv:2410.04844, 2024

    Feng Tian, Yixuan Li, Yichao Yan, Shanyan Guan, Yan- hao Ge, and Xiaokang Yang. Postedit: Posterior sam- pling for efficient zero-shot image editing.arXiv preprint arXiv:2410.04844, 2024. 8

    2024 arXiv

  36. [44]

    Plug-and-play diffusion features for text-driven image-to- image translation

    Narek Tumanyan, Michal Geyer, Shai Bagon, and Tali Dekel. Plug-and-play diffusion features for text-driven image-to- image translation. InCVPR, pages 1921–1930, 2023. 6

    1921

  37. [45]

    Prolificdreamer: High-fidelity and diverse text-to-3d generation with variational score distilla- tion

    Zhengyi Wang, Cheng Lu, Yikai Wang, Fan Bao, Chongxuan Li, Hang Su, and Jun Zhu. Prolificdreamer: High-fidelity and diverse text-to-3d generation with variational score distilla- tion. InNeurIPS, 2024. 1, 3

    2024

  38. [46]

    Godiva: Generating open-domain videos from natural descriptions

    Chenfei Wu, Lun Huang, Qianxi Zhang, Binyang Li, Lei Ji, Fan Yang, Guillermo Sapiro, and Nan Duan. Godiva: Generating open-domain videos from natural descriptions. arXiv preprint arXiv:2104.14806, 2021. 7

    2021 arXiv

  39. [47]

    Gaussctrl: multi-view consistent text-driven 3d gaussian splatting editing

    Jing Wu, Jia-Wang Bian, Xinghui Li, Guangrun Wang, Ian Reid, Philip Torr, and Victor Adrian Prisacariu. Gaussctrl: multi-view consistent text-driven 3d gaussian splatting editing. arXiv preprint arXiv:2403.08733, 2024. 3

    2024 arXiv

  40. [48]

    Inversion-free image editing with natural language

    Sihan Xu, Yidong Huang, Jiayi Pan, Ziqiao Ma, and Joyce Chai. Inversion-free image editing with natural language. In CVPR, 2024. 8

    2024

  41. [49]

    Text-to-3d with classifier score distillation

    Xin Yu, Yuan-Chen Guo, Yangguang Li, Ding Liang, Song- Hai Zhang, and Xiaojuan Qi. Text-to-3d with classifier score distillation. InICLR, 2024. 1, 2, 3, 4, 5, 8

    2024

  42. [50]

    Adding conditional control to text-to-image diffusion models

    Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. Adding conditional control to text-to-image diffusion models. In ICCV, pages 3836–3847, 2023. 3

    2023

  43. [51]

    The unreasonable effectiveness of deep features as a perceptual metric

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shechtman, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. InCVPR, 2018. 7

    2018

  44. [52]

    RePaint-NeRF: Nerf editting via semantic masks and diffusion models

    Xingchen Zhou, Ying He, F Richard Yu, Jianqiang Li, and You Li. RePaint-NeRF: Nerf editting via semantic masks and diffusion models. InIJCAI, 2023. 3

    2023

  45. [53]

    HIFA: High- fidelity text-to-3d generation with advanced diffusion guid- ance

    Junzhe Zhu, Peiye Zhuang, and Sanmi Koyejo. HIFA: High- fidelity text-to-3d generation with advanced diffusion guid- ance. InICLR, 2024. 1, 3 10

    2024

Pith tools

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