Pith. sign in

REVIEW 3 major objections 4 minor 145 references

Diffusion Counterfactual Generation with Semantic Abduction

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read This paper claims that diffusion-based image counterfactuals can preserve identity by abducting a semantic latent code that stays fixed under intervention, at a small cost in effectiveness.

desk verdict Solid empirical contribution with a real but acknowledged identifiability caveat; worth reading for diffusion counterfactual work, though the central identity-preservation claim is not fully settled. read the letter →

arxiv 2506.07883 v1 pith:NGBGEBW7 submitted 2025-06-09 cs.LG cs.AIcs.CVstat.ML

Rajat Rasal , Avinash Kori , Fabio De Sousa Ribeiro , Tian Xia , Ben Glocker This is my paper
classification cs.LGcs.AIcs.CVstat.ML
keywords counterfactualimagegenerationdeepstructuralcausalmodelsdiffusionsemanticabductionidentitypreservationclassifier-freeguidanceinferencemedicalediting
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 tries to establish that diffusion models can edit images counterfactually without silently changing who or what is in the image, by upgrading the abduction step of the structural-causal-model recipe. Instead of only inverting the diffusion trajectory into a spatial noise field, the authors add a variational semantic encoder that extracts a high-level code $z$ from the image and hold that code fixed while the causal parents (the attributes being edited) are intervened upon. On Morpho-MNIST, CelebA-HQ, and EMBED mammograms, semantic abduction improves composition and reversibility---the standard identity-preservation measures---compared with spatial-only abduction at the same guidance settings, while losing only a few points of intervention effectiveness. A dynamic variant, which optimises per-timestep guidance tokens to align the counterfactual trajectory with the inverse trajectory, further preserves backgrounds, facial structure, and illumination. This matters because faithful, identity-preserving counterfactuals are what make fairness audits, medical artefact removal, and causal explanations trustworthy.

What carries the argument

The load-bearing object is the decomposed exogenous noise $\epsilon=(u,z)$ with the approximate posterior $p_S(\epsilon|x,pa)\approx q_\phi(z|e_\phi(x))\,\delta(u-h^{-1}_\theta(x,(z,pa)))$, together with the amortised anti-causal guidance update $\epsilon_\theta(x_t,\emptyset,t)+\omega(\epsilon_\theta(x_t,\tilde{c}_{sem},t)-\epsilon_\theta(x_t,\emptyset,t))$. The spatial term $u$ encodes low-level structure via DDIM inversion; the semantic term $z$ is the decodable invariant content; dynamic abduction adds per-timestep guidance tokens $\emptyset^*_t$ optimised by counterfactual trajectory alignment. The mechanism carries the argument because holding $z$ fixed is exactly what localises the edit to the intervened parents, while the sharpened anti-causal score enforces effectiveness.

What would settle it

Train a probe to predict the intervened attribute from the abducted semantic code $z$ alone; if the probe recovers the attribute substantially above chance on images where the intervention changes only the parents, the semantic code is entangled with the parents and the method's core assumption fails.

Watch

Extended reading notes

Core claim

The central discovery is a clean split of diffusion exogenous noise into a spatial term and a semantic term, with the semantic term drawn from a learned amortised posterior and kept invariant under intervention. Formally, the image mechanism is written as $x := f_\theta(\epsilon, pa) \approx h^\omega_\theta(u, (z, pa))$, where $u$ is obtained by DDIM inversion of $x$ given $(z, pa)$, and $z \sim q_\phi(z|e_\phi(x))$ is a variational code learned end-to-end with a $\beta$-KL regulariser. At intervention time the parents change from $pa$ to $\tilde{pa}$ while $z$ stays fixed, so the generated counterfactual continues to share the high-level content of the original. The paper's headline evidence is Table 3: at $\omega=2$, $p_\emptyset=0.1$, the semantic mechanism reaches LPIPS 0.096 for eyeglasses against 0.171 for the spatial mechanism, with F1 effectiveness staying within a few points; Figure 3c shows that dynamic semantic abduction additionally preserves background, hairstyle, skin tone and facial structure.

Load-bearing premise

The whole method rests on the assumption that the semantic code $z$ carries only identity information, not the attribute being edited; if the code secretly contains the intervened attribute, holding it fixed will leak the original attribute value into the counterfactual.

Editorial extensions

If this is right

  • At fixed guidance settings, semantic mechanisms match spatial mechanisms on intervention faithfulness (F1 or accuracy within a few points) while roughly halving perceptual identity loss, so the identity/effectiveness trade-off becomes a controllable operating point rather than a fixed property of diffusion.
  • Increasing the guidance scale $\omega$ improves effectiveness but monotonically worsens composition and reversibility in both mechanisms, so users can choose where to sit on the curve.
  • Dynamic semantic abduction with counterfactual trajectory alignment improves preservation of background, hairstyle, skin tone, and facial structure; on EMBED it removes 95.16% of triangular and 91.69% of circular skin markers while holding density and cancer labels fixed.
  • Diffusion models tuned for diverse random sampling (large $\omega$, small $p_\emptyset$) turn out to be the wrong operating regime for counterfactual soundness, so L3-style causal evaluation should guide conditioning design.

Reading between the lines

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

  • Beyond the paper, the same semantic-abduction idea should be testable in text-guided latent diffusion: a fixed identity embedding could play the role of $z$, holding the subject stable while a prompt changes the attribute.
  • A direct probe experiment---train a classifier to predict the intervened attribute from $z$ alone on held-out images---would quantify how much of the identity-preservation result depends on the exogeneity assumption the paper leaves unverified.
  • The reported trade-off curves suggest that counterfactual benchmarks should report soundness as a function of $\omega$ and $p_\emptyset$ rather than at a single operating point, because a single point can rank methods almost arbitrarily.
  • Counterfactual trajectory alignment could transfer to image-conditional editing beyond null-text inversion, potentially improving reversibility in any guided diffusion editor that optimises per-step tokens.
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

3 major / 4 minor

Summary. The paper introduces a suite of diffusion-based structural causal mechanisms for counterfactual image generation, framed through Pearl's abduction-action-prediction procedure. It proposes (i) a spatial mechanism using conditional DDIM inversion for abduction, (ii) a semantic mechanism that adds a variational encoder producing a latent z treated as an exogenous noise term and held fixed under interventions, and (iii) dynamic semantic abduction with counterfactual trajectory alignment (CTA), which optimizes guidance tokens to align inverse and guided trajectories. The methods are evaluated on Morpho-MNIST, CelebA-HQ, and EMBED mammography data using axiomatic soundness metrics (composition, reversibility, effectiveness) and LPIPS-based identity preservation, with comparisons against VAE, HVAE, VCI, and DiffSCM baselines. The central claim is that semantic abduction improves identity preservation over purely spatial abduction at a modest cost in intervention effectiveness, and that dynamic ablation further preserves backgrounds and facial structure.

Significance. If the claims hold, the paper is a useful empirical contribution: it extends diffusion-based deep SCMs with a high-level semantic latent, introduces a practical test-time optimization for counterfactual trajectory alignment, and evaluates on three datasets with established external metrics and reported standard deviations in the main CelebA-HQ table. The writing is clear and the limitations are honestly stated. However, the central interpretation depends on an exogeneity assumption for z that the paper explicitly does not guarantee, and the main quantitative tables do not include the dynamic mechanism that is presented as a headline contribution. These issues are concrete and addressable within the paper's scope, so the appropriate disposition is major revision rather than acceptance in the current form.

major comments (3)
  1. [§3.2 and Limitations] The load-bearing claim that semantic abduction enables a 'principled trade-off' between causal control and identity preservation rests on treating z as exogenous noise independent of pa in Equations (12)-(14). The paper explicitly states in the Limitations that 'we do not guarantee that pa⊥⊥z|x'. With the non-identifiable unconditional prior p(z)=N(0,I), there is no mechanism preventing z from encoding the intervened attribute. If z encodes 'eyeglasses' or 'smiling', holding z fixed during do(g) or do(s) can carry the original attribute into the counterfactual, simultaneously lowering LPIPS and lowering effectiveness. Table 3 is consistent with this alternative explanation: for do(g) at ω=2, p∅=0.1, semantic obtains F1(g)=96.86 vs. spatial 99.07 while LPIPS improves from 0.171 to 0.096. The manuscript therefore does not rule out the possibility that the reported identity gain is an artifact of attribute leakage rather than a genuine causal property. A concrete fix within scope would be to train a probe to predict pa from abducted z, to ablate by re-sampling z from the prior p(z) during prediction, or to condition the semantic prior on pa and re-run Table 3.
  2. [§4, Table 3 and Figure 3c/5] Dynamic semantic abduction is presented as a headline contribution in Section 3.3, and the text claims it further improves preservation of backgrounds, hairstyle, skin tone, and facial structure. However, Table 3 contains no rows for 'Semantic w/ Dynamic Abd.', and Figure 5 only shows IDP as a function of η for dynamic abduction, without a matched static-semantic baseline in the same plot. The quantitative evidence for the dynamic claim is therefore limited to qualitative panels in Figure 3c. Please add a quantitative soundness table (composition, reversibility, IDP, effectiveness) for dynamic semantic abduction under matched ω, p∅, and η, including an η=0 no-update control.
  3. [Table 3, do(s) block] The abstract and Section 4 characterize the effectiveness cost of semantic abduction as 'minor' or 'small'. For the smiling intervention at ω=2, p∅=0.1, semantic obtains F1(s)=94.93 vs. spatial F1(s)=99.09, a drop of 4.16 points on the primary effectiveness metric, while IDP improves from 0.139 to 0.066. This is a substantial cost on the direct intervention target and should be discussed as such; the current wording understates a result that is central to the claimed trade-off.
minor comments (4)
  1. [Equation (13)] The expectation in the training objective is written as E_{x,c,t,ε,z} without specifying the distribution of z; please clarify that z is sampled from the variational posterior q_φ(z|e_φ(x)), since this matters for the correct interpretation of the objective.
  2. [Equation (20) and Algorithm 1] The notation x_{t-1}=h^{-1}_{t-1|C}(x_{t-2}) is inconsistent with Equation (7), where h^{-1}_{t|C} maps x_{t-1} to x_t, and with Algorithm 1, which takes the already-inverted trajectory as input. Please clarify whether x_{t-1} is recomputed from x_{t-2} or read from the stored inversion trajectory.
  3. [References] The citation 'Anonymous, 2024' appears in the Introduction and in the reference list as a submission under review; a published proceedings paper should not cite an anonymous under-review manuscript, so this should be replaced with a citable source or removed.
  4. [Tables 1 and 2] Standard deviations are reported for the CelebA-HQ results in Table 3 but not for the Morpho-MNIST results in Tables 1 and 2, making it difficult to assess whether small differences, such as the composition values separating spatial and semantic mechanisms, are meaningful.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular derivation found; the paper is an empirical methods paper with an acknowledged identifiability caveat that could confound, but does not define, its central comparison.

full rationale

This is an empirical methods paper, not a derivation chain, and its central claims are supported by external benchmark metrics rather than by construction. Semantic abduction (Eqs. 12-15) holds a latent z fixed during intervention; the paper explicitly disclaims the key independence condition in the Limitations: 'we do not guarantee that pa⊥⊥z|x, but instead set the dimensionality of z and value of β such that we can control pa whilst improving identity preservation.' This is an identifiability and confounding caveat that could explain part of the reported IDP gain as attribute leakage, but it is not circular: IDP is not defined in terms of z, and no fitted parameter is renamed as a prediction. Counterfactual trajectory alignment (Eq. 20) optimizes guidance tokens to minimize trajectory mismatch, so the dynamic-abduction improvements partly reflect the optimization objective, but the headline semantic-vs-spatial comparison at matched ω and p∅ is not generated by that objective. Self-citations (e.g., Monteiro et al. 2023 for soundness metrics, De Sousa Ribeiro et al. 2023 as baseline) are standard prior work and are not used to force the central claim. I found no step where Eq. X equals Eq. Y by construction or where a parameter fitted to the target metric is presented as a prediction.

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

No new physical entities, forces, or mediators are introduced. The semantic latent z is a standard variational latent variable, and the guidance tokens are per-step optimization variables. The central claim instead rests on modeling assumptions (Markovian SCM, bijective DDIM inversion, diagonal-Gaussian amortised posterior) and on several hand-tuned hyperparameters (omega, p_empty, eta, beta, z size).

free parameters (5)
  • guidance scale omega = 1.2 to 4.5 (e.g., 1.5, 2, 3)
    Controls strength of amortised anti-causal guidance; swept across experiments to demonstrate the effectiveness versus identity trade-off (Tables 1 to 3, Figure 4).
  • conditioning dropout probability p_empty = 0.1, 0.2, 0.5
    Probability of replacing the condition with the empty token during training for classifier-free guidance; the authors choose p_empty for Level-3 soundness rather than Level-2 diversity (Section 3.3).
  • CTA step size eta = 0.001 to 0.01 (fine-tuned per observation in Figure 3c)
    Update step for counterfactual trajectory alignment; the caption of Figure 3c states the choice is fine-tuned for each observation, so the qualitative results use per-image tuning.
  • semantic KL weight beta = not reported
    Weight on the KL divergence in Equation (13); the authors state they set the dimensionality of z and beta to balance causal control and identity preservation (Limitations), but no value or tuning procedure is given.
  • semantic latent dimensionality z = 8 (Morpho-MNIST), 32 (CelebA-HQ), 512 (EMBED)
    Chosen by hand per dataset to balance causal control and identity preservation (Table 4).
assumptions (5)
  • domain assumption Markovian SCM with acyclic, invertible mechanisms and independent exogenous noise
    Assumed in Section 2.1 to set up abduction-action-prediction; the authors note in Limitations that this need not hold in practice.
  • domain assumption DDIM inversion is a bijection under the null intervention with a perfect noise estimator
    Invoked in Section 3.1 to justify spatial abduction with the Dirac delta posterior on u.
  • ad hoc to paper The semantic posterior is approximated by a diagonal Gaussian q_phi(z|e_phi(x)), and z is treated as independent exogenous noise, p(z)=N(0,I)
    Defined in Section 3.2, Equations (13) and (14); the model is non-identifiable and pa independent of z given x is not guaranteed (Limitations).
  • domain assumption The causal graphs for CelebA-HQ (s implies m, o as independent parent of x) and EMBED (t, c, d, y as independent parents) are correct
    Assumed in Section 4 and Section I; the authors add confounder o after observing spurious correlations such as hair loss and glasses removal (Section 4, CelebA-HQ paragraph).
  • domain assumption The denoiser can amortise conditional and unconditional representations for classifier-free guidance
    Needed for Equation (16) amortised anti-causal guidance; standard classifier-free guidance assumption from Ho and Salimans (2022).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Diffusion Counterfactual Generation with Semantic Abduction." pith.science (2026). https://pith.science/paper/NGBGEBW7

@misc{pith2026250607883,
  author       = {Pith},
  title        = {Pith review of: Diffusion Counterfactual Generation with Semantic Abduction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NGBGEBW7}},
  note         = {Machine review of arXiv:2506.07883}
}
read the original abstract

Counterfactual image generation presents significant challenges, including preserving identity, maintaining perceptual quality, and ensuring faithfulness to an underlying causal model. While existing auto-encoding frameworks admit semantic latent spaces which can be manipulated for causal control, they struggle with scalability and fidelity. Advancements in diffusion models present opportunities for improving counterfactual image editing, having demonstrated state-of-the-art visual quality, human-aligned perception and representation learning capabilities. Here, we present a suite of diffusion-based causal mechanisms, introducing the notions of spatial, semantic and dynamic abduction. We propose a general framework that integrates semantic representations into diffusion models through the lens of Pearlian causality to edit images via a counterfactual reasoning process. To our knowledge, this is the first work to consider high-level semantic identity preservation for diffusion counterfactuals and to demonstrate how semantic control enables principled trade-offs between faithful causal control and identity preservation.

Figures

Figures reproduced from arXiv: 2506.07883 by the authors.

Figure 1
Figure 1. Twin network representations for our diffusion mechanisms. Black and white arrowheads refer resp. to the generative and abductive / inference directions. Edges ending in black circles depict conditions. Circular and diamond nodes refer resp. to depict random and deterministic variables. Boxes house the independent exogenous decomposition of ϵ. In (a)-(b), generation and inference are performed resp. with DDIM hθ(·) … view at source ↗
Figure 2
Figure 2. Morpho-MNIST (28 × 28) counterfactuals generated using an amortised, anti-causally guided semantic mechanism (p∅ = 0.1, ω = 1.5) based on the DSCM shown in (a). (b) illustrates counterfactual soundness (Obs: Observation, Comp: Composition, Cf: Counterfactual, Rev: Reversibility). (c) depicts image counterfactuals: interventions are shown above the top row and the bottom row visualises total causal effects (red: incr… view at source ↗
Figure 3
Figure 3. CelebA-HQ (64 × 64) counterfactuals generated using amortised, anti-causally guided semantic mechanisms. (a) DSCM with spatial (u), semantic (z) and dynamic (∅ ∗ 1:T ) exogenous noise terms for x. (b) shows counterfactual soundness using semantic abduction with p∅ = 0.1 and ω = 2. (c) shows that semantic mechanisms improve identity preservation, and dynamic abduction further improves backgrounds, hairstyle, skin col… view at source ↗
Figures from the paper (13 more)
Figure 5
Figure 5. Figure 5: CelebA-HQ: Effect of step size (η) and guidance scale (ω) on model identity preservation (IDP) when using dynamic ab￾duction with amortised anti-causally guided semantic mechanism (p∅ = 0.1) on 200 randomly chosen images from the val. set. improvement in composition, w…
Figure 6
Figure 6. Figure 6: shows that our mechanisms effectively remove arte￾facts and can disentangle representations for triangles and circles. We successfully remove 95.16 ± 1.34% of triangles and 91.69 ± 1.06% of circles in our test set - a notewor￾thy result given the dataset’s small size a…
Figure 7
Figure 7. Figure 7: Examples from Morpho-MNIST 0 1 2 3 4 5 6 7 8 9 Digit Thickness Slant Width Intensity 2 4 6 All -25° 0° 25° 10 20 100 200 [PITH_FULL_IMAGE:figures/full_fig_p022_7.png]
Figure 9
Figure 9. Figure 9: Morpho-MNIST (28 × 28) counterfactuals generated using an amortised, anti-causally guided semantic mechanism (p∅ = 0.1, ω = 1.5) based on the DSCM shown in Figure 2a. Interventions are shown above the top row and the bottom row visualises total causal effects (red: inc…
Figure 10
Figure 10. Figure 10: Morpho-MNIST (28 × 28) counterfactuals generated using an amortised, anti-causally guided spatial mechanism (p∅ = 0.1, ω = 1.5) based on the DSCM shown in Figure 2a. Interventions are shown above the top row and the bottom row visualises total causal effects (red: inc…
Figure 11
Figure 11. Figure 11: Examples from colour Morpho-MNIST 0 1 2 3 4 5 6 7 8 9 Digit Thickness Slant Hue 2 4 6 All -25° 0° 25° 0.0 0.5 1.0 [PITH_FULL_IMAGE:figures/full_fig_p025_11.png]
Figure 13
Figure 13. Figure 13: Colourised Morpho-MNIST (28 × 28) counterfactuals generated using an amortised, anti-causally guided semantic mechanism (p∅ = 0.1, ω = 1.5). (a) depicts the DSCM: h is hue, d is digit class, s is slant, t is thickness and x is the image. (b) depicts image counterfactu…
Figure 14
Figure 14. Figure 14: CelebA (64 × 64) counterfactuals generated using our amortised, anti-causally guided mechanisms in the DSCM in Figure 3a. H. CelebA-HQ H.1. Additional Results d o(g = 1) Observation Composition Counterfactual Reversal d o(g = 0) d o(s = 1) d o(s = 0) d o(g = 1) Observ…
Figure 15
Figure 15. Figure 15: CelebA-HQ (64 × 64) counterfactuals using our amortised, anti-causally guided mechanisms in the DSCM in Figure 3a with ω = 2. Notice that identity preservation improves in each figure from left to right. d o(s = 0) Observation Spatial p = 0.1 Semantic p = 0.1 Semantic…
Figure 16
Figure 16. Figure 16: CelebA-HQ (64 × 64) counterfactuals generated using amortised, anti-causally guided mechanisms. Notice that identity preservation improves as we use increasingly complex semantic abduction procedures and increase the value of p∅. 26 [PITH_FULL_IMAGE:figures/full_fig_…
Figure 17
Figure 17. Figure 17: EMBED (192 × 192) counterfactuals using an amortised, anti-causally guided semantic mechanism with (p∅ = 0.1, ω = 1.2) for circular skin marker removal. We find that larger ω improves effectiveness whilst compromising illumination and breast density. Remove Triangles …
Figure 18
Figure 18. Figure 18: EMBED (192 × 192) counterfactuals using an amortised, anti-causally guided semantic mechanism with (p∅ = 0.1, ω = 1.2) for triangular skin marker removal. We find that larger ω improves effectiveness whilst compromising illumination and breast density. Remove Triangle…
Figure 19
Figure 19. Figure 19: EMBED (192 × 192) counterfactuals using an amortised, anti-causally guided semantic mechanism with (p∅ = 0.1, ω = 1.2) with dynamic abduction (η = 0.001) for triangular and circular marker removal. Counterfactual inference takes ∼ 5 mins per image. In these cases, dyn…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

145 extracted references · 31 canonical work pages

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION format.date year duplicate empty "emp...

  2. [2]

    Diffusion-based representation learning

    Abstreiter, K., Mittal, S., Bauer, S., Sch \"o lkopf, B., and Mehrjou, A. Diffusion-based representation learning. arXiv preprint arXiv:2105.14257, 2021

    arXiv 2021

  3. [3]

    Scaling in-the-wild training for diffusion-based illumination harmonization and editing by imposing consistent light transport

    Anonymous. Scaling in-the-wild training for diffusion-based illumination harmonization and editing by imposing consistent light transport. In Submitted to The Thirteenth International Conference on Learning Representations, 2024. URL https://openreview.net/forum?id=u1cQYxRI1H. under review

    2024

  4. [4]

    Counterfactual Explanations for Medical Image Classification and Regression using Diffusion Autoencoder

    Atad, M., Schinz, D., Moeller, H., Graf, R., Wiestler, B., Rueckert, D., Navab, N., Kirschke, J. S., and Keicher, M. Counterfactual explanations for medical image classification and regression using diffusion autoencoder. arXiv preprint arXiv:2408.01571, 2024

    work page Pith review arXiv 2024

  5. [5]

    Diffusion visual counterfactual explanations

    Augustin, M., Boreiko, V., Croce, F., and Hein, M. Diffusion visual counterfactual explanations. Advances in Neural Information Processing Systems, 35: 0 364--377, 2022

    2022

  6. [6]

    D., Ibeling, D., and Icard, T

    Bareinboim, E., Correa, J. D., Ibeling, D., and Icard, T. On Pearl’s Hierarchy and the Foundations of Causal Inference, pp.\ 507–556. Association for Computing Machinery, New York, NY, USA, 1 edition, 2022. ISBN 9781450395861. URL https://doi.org/10.1145/3501714.3501743

    arXiv 2022

  7. [7]

    Variational diffusion auto-encoder: Latent space extraction from pre-trained diffusion models

    Batzolis, G., Stanczuk, J., and Sch \"o nlieb, C.-B. Variational diffusion auto-encoder: Latent space extraction from pre-trained diffusion models. arXiv preprint arXiv:2304.12141, 2023

    arXiv 2023

  8. [8]

    P., Higgins, I., Pal, A., Matthey, L., Watters, N., Desjardins, G., and Lerchner, A

    Burgess, C. P., Higgins, I., Pal, A., Matthey, L., Watters, N., Desjardins, G., and Lerchner, A. Understanding disentangling in beta- VAE . arXiv preprint arXiv:1804.03599, 2018

    arXiv 2018

Show all 145 references
  1. [9]

    C., Tan, J., Kainz, B., Konukoglu, E., and Glocker, B

    Castro, D. C., Tan, J., Kainz, B., Konukoglu, E., and Glocker, B. Morpho-MNIST : Quantitative assessment and diagnostics for representation learning. Journal of Machine Learning Research, 20 0 (178), 2019

    2019

  2. [10]

    Chao, P., Bl \"o baum, P., and Kasiviswanathan, S. P. Interventional and counterfactual inference with diffusion models. arXiv preprint arXiv:2302.00860, 2023

    2023 arXiv

  3. [11]

    Generalizable single-source cross-modality medical image segmentation via invariant causal mechanisms

    Chen, B., Zhu, Y., Ao, Y., Caprara, S., Sutter, R., R \"a tsch, G., Konukoglu, E., and Susmelj, A. Generalizable single-source cross-modality medical image segmentation via invariant causal mechanisms. In 2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV...

    2025

  4. [12]

    Q., Li, X., Grosse, R., and Duvenaud, D

    Chen, T. Q., Li, X., Grosse, R., and Duvenaud, D. Isolating sources of disentanglement in variational autoencoders. In Advances in Neural Information Processing Systems, 2018

    2018

  5. [13]

    Infogan: Interpretable representation learning by information maximizing generative adversarial nets

    Chen, X., Duan, Y., Houthooft, R., Schulman, J., Sutskever, I., and Abbeel, P. Infogan: Interpretable representation learning by information maximizing generative adversarial nets. Advances in neural information processing systems, 29, 2016

    2016

  6. [14]

    Very deep vaes generalize autoregressive models and can outperform them on images

    Child, R. Very deep vaes generalize autoregressive models and can outperform them on images. arXiv preprint arXiv:2011.10650, 2020

    2011 arXiv

  7. [15]

    K., Lu, J., Inouye, D

    Cho, W., Ravi, H., Harikumar, M., Khuc, V., Singh, K. K., Lu, J., Inouye, D. I., and Kale, A. Towards enhanced controllability of diffusion models. arXiv preprint arXiv:2302.14368, 2023

    2023 arXiv

  8. [16]

    Y., Nam, H., and Ye, J

    Chung, H., Kim, J., Park, G. Y., Nam, H., and Ye, J. C. Cfg++: Manifold-constrained classifier free guidance for diffusion models. arXiv preprint arXiv:2406.08070, 2024

    2024 arXiv

  9. [17]

    and Jaini, P

    Clark, K. and Jaini, P. Text-to-image diffusion models are zero shot classifiers. Advances in Neural Information Processing Systems, 36, 2024

    2024

  10. [18]

    Diffedit: Diffusion-based semantic image editing with mask guidance

    Couairon, G., Verbeek, J., Schwenk, H., and Cord, M. Diffedit: Diffusion-based semantic image editing with mask guidance. arXiv preprint arXiv:2210.11427, 2022

    2022 arXiv

  11. [19]

    N., and Sharma, A

    Dash, S., Balasubramanian, V. N., and Sharma, A. Evaluating and mitigating bias in image classifiers: A causal perspective using counterfactuals. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp.\ 915--924, 2022

    2022

  12. [20]

    High fidelity image counterfactuals with probabilistic causal models

    De Sousa Ribeiro, F., Xia, T., Monteiro, M., Pawlowski, N., and Glocker, B. High fidelity image counterfactuals with probabilistic causal models. In Proceedings of the 40th International Conference on Machine Learning, pp.\ 7390--7425, 2023

    2023

  13. [21]

    and Nichol, A

    Dhariwal, P. and Nichol, A. Diffusion models beat gans on image synthesis. Advances in neural information processing systems, 34: 0 8780--8794, 2021

    2021

  14. [22]

    Rethinking conditional diffusion sampling with progressive guidance

    Dinh, A.-D., Liu, D., and Xu, C. Rethinking conditional diffusion sampling with progressive guidance. In Oh, A., Naumann, T., Globerson, A., Saenko, K., Hardt, M., and Levine, S. (eds.), Advances in Neural Information Processing Systems, volume 36, pp.\ 42285--42297. Curran As...

    2023

  15. [23]

    Diffusion self-guidance for controllable image generation

    Epstein, D., Jabri, A., Poole, B., Efros, A., and Holynski, A. Diffusion self-guidance for controllable image generation. Advances in Neural Information Processing Systems, 36: 0 16222--16239, 2023

    2023

  16. [24]

    Diffexplainer: Unveiling black box models via counterfactual generation

    Fang, Y., Wu, S., Jin, Z., Wang, S., Xu, C., Walsh, S., and Yang, G. Diffexplainer: Unveiling black box models via counterfactual generation. In International Conference on Medical Image Computing and Computer-Assisted Intervention, pp.\ 208--218. Springer, 2024

    2024

  17. [25]

    H., Chechik, G., and Cohen-Or, D

    Gal, R., Alaluf, Y., Atzmon, Y., Patashnik, O., Bermano, A. H., Chechik, G., and Cohen-Or, D. An image is worth one word: Personalizing text-to-image generation using textual inversion. arXiv preprint arXiv:2208.01618, 2022

    2022 arXiv

  18. [26]

    and Pearl, J

    Galles, D. and Pearl, J. An axiomatic characterization of causal counterfactuals. Foundations of Science, 3: 0 151--182, 1998

    1998

  19. [27]

    Geirhos, R., Jacobsen, J.-H., Michaelis, C., Zemel, R., Brendel, W., Bethge, M., and Wichmann, F. A. Shortcut learning in deep neural networks. Nature Machine Intelligence, 2 0 (11): 0 665--673, 2020

    2020

  20. [28]

    Generative adversarial nets

    Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y. Generative adversarial nets. Advances in neural information processing systems, 27, 2014

    2014

  21. [29]

    P., Gao, J., and Poon, H

    Gu, Y., Yang, J., Usuyama, N., Li, C., Zhang, S., Lungren, M. P., Gao, J., and Poon, H. Biomedjourney: Counterfactual biomedical image generation by instruction-learning from multimodal patient journeys. arXiv preprint arXiv:2310.10765, 2023

    arXiv

  22. [30]

    S., and Michaeli, T

    Haas, R., Huberman-Spiegelglas, I., Mulayoff, R., Gra hof, S., Brandt, S. S., and Michaeli, T. Discovering interpretable directions in the semantic latent space of diffusion models. In 2024 IEEE 18th International Conference on Automatic Face and Gesture Recognition (FG), pp.\...

    2024

  23. [31]

    Halpern, J. Y. Axiomatizing causal reasoning. Journal of Artificial Intelligence Research, 12: 0 317--337, 2000

    2000

  24. [32]

    Prompt-to-prompt image editing with cross attention control

    Hertz, A., Mokady, R., Tenenbaum, J., Aberman, K., Pritch, Y., and Cohen-Or, D. Prompt-to-prompt image editing with cross attention control. arXiv preprint arXiv:2208.01626, 2022

    2022 arXiv

  25. [33]

    beta- VAE : Learning basic visual concepts with a constrained variational framework

    Higgins, I., Matthey, L., Pal, A., Burgess, C., Glorot, X., Botvinick, M., Mohamed, S., and Lerchner, A. beta- VAE : Learning basic visual concepts with a constrained variational framework. In International Conference on Learning Representations, 2017. URL https://openreview.n...

    2017

  26. [34]

    and Salimans, T

    Ho, J. and Salimans, T. Classifier-free diffusion guidance. arXiv preprint arXiv:2207.12598, 2022

    2022 arXiv

  27. [35]

    Denoising diffusion probabilistic models

    Ho, J., Jain, A., and Abbeel, P. Denoising diffusion probabilistic models. Advances in neural information processing systems, 33: 0 6840--6851, 2020

    2020

  28. [36]

    Enforcing conditional independence for fair representation learning and causal image generation

    Hwa, J., Zhao, Q., Lahiri, A., Masood, A., Salimi, B., and Adeli, E. Enforcing conditional independence for fair representation learning and causal image generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.\ 103--112, 2024

    2024

  29. [37]

    and Morioka, H

    Hyvarinen, A. and Morioka, H. Unsupervised feature extraction by time-contrastive learning and nonlinear ica. Advances in neural information processing systems, 29, 2016

    2016

  30. [38]

    Nonlinear ica using auxiliary variables and generalized contrastive learning

    Hyvarinen, A., Sasaki, H., and Turner, R. Nonlinear ica using auxiliary variables and generalized contrastive learning. In The 22nd International Conference on Artificial Intelligence and Statistics, pp.\ 859--868. PMLR, 2019

    2019

  31. [39]

    Semi-supervised learning for deep causal generative models

    Ibrahim, Y., Warr, H., and Kamnitsas, K. Semi-supervised learning for deep causal generative models. arXiv preprint arXiv:2403.18717, 2024

    2024 arXiv

  32. [40]

    Intriguing properties of generative classifiers

    Jaini, P., Clark, K., and Geirhos, R. Intriguing properties of generative classifiers. arXiv preprint arXiv:2309.16779, 2023

    2023 arXiv

  33. [41]

    J., Vey, B

    Jeong, J. J., Vey, B. L., Reddy, A., Kim, T., Santos, T., Correa, R., Dutt, R., Mosunjac, M., Oprea-Ilies, G., Smith, G., et al. The emory breast imaging dataset (embed): a racially diverse, granular dataset of 3.5 m screening and diagnostic mammograms. arXiv preprint arXiv:22...

    2022 arXiv

  34. [42]

    Progressive growing of gans for improved quality, stability, and variation

    Karras, T. Progressive growing of gans for improved quality, stability, and variation. arXiv preprint arXiv:1710.10196, 2017

    2017 arXiv

  35. [43]

    A style-based generator architecture for generative adversarial networks

    Karras, T. A style-based generator architecture for generative adversarial networks. arXiv preprint arXiv:1812.04948, 2019

    2019 arXiv

  36. [44]

    Analyzing and improving the image quality of stylegan

    Karras, T., Laine, S., Aittala, M., Hellsten, J., Lehtinen, J., and Aila, T. Analyzing and improving the image quality of stylegan. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.\ 8110--8119, 2020

    2020

  37. [45]

    Guiding a diffusion model with a bad version of itself

    Karras, T., Aittala, M., Kynk \"a \"a nniemi, T., Lehtinen, J., Aila, T., and Laine, S. Guiding a diffusion model with a bad version of itself. arXiv preprint arXiv:2406.02507, 2024

    2024 arXiv

  38. [46]

    Variational autoencoders and nonlinear ica: A unifying framework

    Khemakhem, I., Kingma, D., Monti, R., and Hyvarinen, A. Variational autoencoders and nonlinear ica: A unifying framework. In International conference on artificial intelligence and statistics, pp.\ 2207--2217. PMLR, 2020 a

    2020

  39. [47]

    Ice-beem: Identifiable conditional energy-based deep models based on nonlinear ica

    Khemakhem, I., Monti, R., Kingma, D., and Hyvarinen, A. Ice-beem: Identifiable conditional energy-based deep models based on nonlinear ica. Advances in Neural Information Processing Systems, 33: 0 12768--12778, 2020 b

    2020

  40. [48]

    and Mnih, A

    Kim, H. and Mnih, A. Disentangling by factorising. In International Conference on Machine Learning, 2018

    2018

  41. [49]

    Auto-encoding variational bayes

    Kingma, D. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114, 2013

    2013 arXiv

  42. [50]

    Kingma, D. P. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014

    2014 arXiv

  43. [51]

    u gelgen, J., Sch \

    Kladny, K.-R., von K \"u gelgen, J., Sch \"o lkopf, B., and Muehlebach, M. Deep backtracking counterfactuals for causally compliant explanations. arXiv preprint arXiv:2310.07665, 2023

    2023 arXiv

  44. [52]

    G., and Vishwanath, S

    Kocaoglu, M., Snyder, C., Dimakis, A. G., and Vishwanath, S. Causalgan: Learning causal implicit generative models with adversarial training. arXiv preprint arXiv:1709.02023, 2017

    2017 arXiv

  45. [53]

    From identifiable causal representations to controllable counterfactual generation: A survey on causal generative modeling

    Komanduri, A., Wu, X., Wu, Y., and Chen, F. From identifiable causal representations to controllable counterfactual generation: A survey on causal generative modeling. arXiv preprint arXiv:2310.11011, 2023

    2023 arXiv

  46. [54]

    Causal diffusion autoencoders: Toward counterfactual generation via diffusion probabilistic models

    Komanduri, A., Zhao, C., Chen, F., and Wu, X. Causal diffusion autoencoders: Toward counterfactual generation via diffusion probabilistic models. arXiv preprint arXiv:2404.17735, 2024

    2024 arXiv

  47. [55]

    Variational inference of disentangled latent concepts from unlabeled observations

    Kumar, A., Sattigeri, P., and Balakrishnan, A. Variational inference of disentangled latent concepts from unlabeled observations. In International Conference on Learning Representations, 2018

    2018

  48. [56]

    Li, S., Hooi, B., and Lee, G. H. Identifying through flows for recovering latent representations. arXiv preprint arXiv:1909.12555, 2019

    1909 arXiv

  49. [57]

    Common diffusion noise schedules and sample steps are flawed

    Lin, S., Liu, B., Li, J., and Yang, X. Common diffusion noise schedules and sample steps are flawed. In Proceedings of the IEEE/CVF winter conference on applications of computer vision, pp.\ 5404--5411, 2024

    2024

  50. [58]

    R., and Hand, E

    Lingenfelter, B., Davis, S. R., and Hand, E. M. A quantitative analysis of labeling issues in the celeba dataset. In Advances in Visual Computing: 17th International Symposium, ISVC 2022, San Diego, CA, USA, October 3–5, 2022, Proceedings, Part I, pp.\ 129–141, Berlin, Heidelb...

    2022 doi

  51. [59]

    Deep learning face attributes in the wild

    Liu, Z., Luo, P., Wang, X., and Tang, X. Deep learning face attributes in the wild. In Proceedings of International Conference on Computer Vision (ICCV), December 2015

    2015

  52. [60]

    Challenging common assumptions in the unsupervised learning of disentangled representations

    Locatello, F., Bauer, S., Lucic, M., Raetsch, G., Gelly, S., Sch \"o lkopf, B., and Bachem, O. Challenging common assumptions in the unsupervised learning of disentangled representations. In international conference on machine learning, pp.\ 4114--4124. PMLR, 2019

    2019

  53. [61]

    a tsch, G., Gelly, S., Sch \

    Locatello, F., Bauer, S., Lucic, M., R \"a tsch, G., Gelly, S., Sch \"o lkopf, B., and Bachem, O. A sober look at the unsupervised learning of disentangled representations and their evaluation. Journal of Machine Learning Research, 21 0 (209): 0 1--62, 2020

    2020

  54. [62]

    Understanding diffusion models: A unified perspective

    Luo, C. Understanding diffusion models: A unified perspective. arXiv preprint arXiv:2208.11970, 2022

    2022 arXiv

  55. [63]

    Melistas, T., Spyrou, N., Gkouti, N., Sanchez, P., Vlontzos, A., Papanastasiou, G., and Tsaftaris, S. A. Benchmarking counterfactual image generation. arXiv preprint arXiv:2403.20287, 2024

    2024 arXiv

  56. [64]

    Sdedit: Guided image synthesis and editing with stochastic differential equations

    Meng, C., He, Y., Song, Y., Song, J., Wu, J., Zhu, J.-Y., and Ermon, S. Sdedit: Guided image synthesis and editing with stochastic differential equations. arXiv preprint arXiv:2108.01073, 2021

    2021 arXiv

  57. [65]

    and Osindero, S

    Mirza, M. and Osindero, S. Conditional generative adversarial nets. arXiv preprint arXiv:1411.1784, 2014

    2014 arXiv

  58. [66]

    Diffusion based representation learning

    Mittal, S., Abstreiter, K., Bauer, S., Sch\" o lkopf, B., and Mehrjou, A. Diffusion based representation learning. In Krause, A., Brunskill, E., Cho, K., Engelhardt, B., Sabato, S., and Scarlett, J. (eds.), Proceedings of the 40th International Conference on Machine Learning, ...

    2023

  59. [67]

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

    Mokady, R., Hertz, A., Aberman, K., Pritch, Y., and Cohen-Or, D. Null-text inversion for editing real images using guided diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.\ 6038--6047, 2023

    2023

  60. [68]

    Monteiro, M., Ribeiro, F. D. S., Pawlowski, N., Castro, D. C., and Glocker, B. Measuring axiomatic soundness of counterfactual image models. arXiv preprint arXiv:2303.01274, 2023

    2023 arXiv

  61. [69]

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

    Nichol, A., Dhariwal, P., Ramesh, A., Shyam, P., Mishkin, P., McGrew, B., Sutskever, I., and Chen, M. Glide: Towards photorealistic image generation and editing with text-guided diffusion models. arXiv preprint arXiv:2112.10741, 2021

    2021 arXiv

  62. [70]

    Nichol, A. Q. and Dhariwal, P. Improved denoising diffusion probabilistic models. In International conference on machine learning, pp.\ 8162--8171. PMLR, 2021

    2021

  63. [71]

    and Bareinboim, E

    Pan, Y. and Bareinboim, E. Counterfactual image editing. arXiv preprint arXiv:2403.09683, 2024

    2024 arXiv

  64. [72]

    Diffusevae: Efficient, controllable and high-fidelity generation from low-dimensional latents

    Pandey, K., Mukherjee, A., Rai, P., and Kumar, A. Diffusevae: Efficient, controllable and high-fidelity generation from low-dimensional latents. arXiv preprint arXiv:2201.00308, 2022

    2022 arXiv

  65. [73]

    J., Mohamed, S., and Lakshminarayanan, B

    Papamakarios, G., Nalisnick, E., Rezende, D. J., Mohamed, S., and Lakshminarayanan, B. Normalizing flows for probabilistic modeling and inference. Journal of Machine Learning Research, 22 0 (57): 0 1--64, 2021

    2021

  66. [74]

    Understanding the latent space of diffusion models through the lens of riemannian geometry

    Park, Y.-H., Kwon, M., Choi, J., Jo, J., and Uh, Y. Understanding the latent space of diffusion models through the lens of riemannian geometry. Advances in Neural Information Processing Systems, 36: 0 24129--24142, 2023

    2023

  67. [75]

    Zero-shot image-to-image translation

    Parmar, G., Kumar Singh, K., Zhang, R., Li, Y., Lu, J., and Zhu, J.-Y. Zero-shot image-to-image translation. In ACM SIGGRAPH 2023 Conference Proceedings, pp.\ 1--11, 2023

    2023

  68. [76]

    Deep structural causal models for tractable counterfactual inference

    Pawlowski, N., Coelho de Castro, D., and Glocker, B. Deep structural causal models for tractable counterfactual inference. Advances in Neural Information Processing Systems, 33: 0 857--869, 2020

    2020

  69. [77]

    Direct and indirect effects

    Pearl, J. Direct and indirect effects. Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence, pp.\ 411--420, 2001

    2001

  70. [78]

    Causality

    Pearl, J. Causality. Cambridge University Press, 2 edition, 2009. doi:10.1017/CBO9780511803161

    2009 doi

  71. [79]

    The seven tools of causal inference, with reflections on machine learning

    Pearl, J. The seven tools of causal inference, with reflections on machine learning. Communications of the ACM, 62 0 (3): 0 54--60, 2019

    2019

  72. [80]

    The hessian penalty: A weak prior for unsupervised disentanglement

    Peebles, W., Peebles, J., Zhu, J.-Y., Efros, A., and Torralba, A. The hessian penalty: A weak prior for unsupervised disentanglement. In Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part VI 16, pp.\ 581--597. Springer, 2020

    2020

  73. [81]

    P., van Breugel, B., Castro, D

    P \'e rez-Garc \' a, F., Bond-Taylor, S., Sanchez, P. P., van Breugel, B., Castro, D. C., Sharma, H., Salvatelli, V., Wetscherek, M. T., Richardson, H., Lungren, M. P., et al. Radedit: stress-testing biomedical vision models via diffusion image editing. arXiv preprint arXiv:23...

    2023 arXiv

  74. [82]

    Elements of Causal Inference: Foundations and Learning Algorithms

    Peters, J., Janzing, D., and Schlkopf, B. Elements of Causal Inference: Foundations and Learning Algorithms. The MIT Press, 2017. ISBN 0262037319

    2017

  75. [83]

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

    Podell, D., English, Z., Lacey, K., Blattmann, A., Dockhorn, T., M \"u ller, J., Penna, J., and Rombach, R. Sdxl: Improving latent diffusion models for high-resolution image synthesis. arXiv preprint arXiv:2307.01952, 2023

    2023 arXiv

  76. [84]

    Learning structural causal models through deep generative models: Methods, guarantees, and challenges

    Poinsot, A., Leite, A., Chesneau, N., S \'e bag, M., and Schoenauer, M. Learning structural causal models through deep generative models: Methods, guarantees, and challenges. arXiv preprint arXiv:2405.05025, 2024

    2024 arXiv

  77. [85]

    Lance: Stress-testing visual models by generating language-guided counterfactual images

    Prabhu, V., Yenamandra, S., Chattopadhyay, P., and Hoffman, J. Lance: Stress-testing visual models by generating language-guided counterfactual images. Advances in Neural Information Processing Systems, 36: 0 25165--25184, 2023

    2023

  78. [86]

    Diffusion autoencoders: Toward a meaningful and decodable representation

    Preechakul, K., Chatthee, N., Wizadwongsa, S., and Suwajanakorn, S. Diffusion autoencoders: Toward a meaningful and decodable representation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.\ 10619--10629, 2022

    2022

  79. [87]

    M., Jordan, M., and Kocaoglu, M

    Rahman, M. M., Jordan, M., and Kocaoglu, M. Conditional generative models are sufficient to sample from any causal effect estimand. arXiv preprint arXiv:2402.07419, 2024

    2024 arXiv

  80. [88]

    C., Pawlowski, N., and Glocker, B

    Rasal, R., Castro, D. C., Pawlowski, N., and Glocker, B. Deep structural causal shape models. In European Conference on Computer Vision, pp.\ 400--432. Springer, 2022

    2022

  81. [89]

    C., Oxtoby, N

    Ravi, D., Alexander, D. C., Oxtoby, N. P., and Initiative, A. D. N. Degenerative adversarial neuroimage nets: generating images that mimic disease progression. In International Conference on Medical Image Computing and Computer-Assisted Intervention, pp.\ 164--172. Springer, 2019

    2019

  82. [90]

    C., Carass, A., and Prince, J

    Reinhold, J. C., Carass, A., and Prince, J. L. A structural causal model for mr images of multiple sclerosis. In Medical Image Computing and Computer Assisted Intervention--MICCAI 2021: 24th International Conference, Strasbourg, France, September 27--October 1, 2021, Proceedin...

    2021

  83. [91]

    Rezende, D. J. and Viola, F. Taming vaes. arXiv preprint arXiv:1810.00597, 2018

    2018 arXiv

  84. [92]

    Ribeiro, F. D. S. and Glocker, B. Demystifying variational diffusion models. arXiv preprint arXiv:2401.06281, 2024

    2024 arXiv

  85. [93]

    On linear identifiability of learned representations

    Roeder, G., Metz, L., and Kingma, D. On linear identifiability of learned representations. In International Conference on Machine Learning, pp.\ 9030--9039. PMLR, 2021

    2021

  86. [94]

    High-resolution image synthesis with latent diffusion models

    Rombach, R., Blattmann, A., Lorenz, D., Esser, P., and Ommer, B. High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.\ 10684--10695, 2022

    2022

  87. [95]

    U-net: Convolutional networks for biomedical image segmentation

    Ronneberger, O., Fischer, P., and Brox, T. 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 III 18, ...

    2015

  88. [96]

    Counterfactual contrastive learning: robust representations via causal image synthesis

    Roschewitz, M., de Sousa Ribeiro, F., Xia, T., Khara, G., and Glocker, B. Counterfactual contrastive learning: robust representations via causal image synthesis. In MICCAI Workshop on Data Engineering in Medical Imaging, pp.\ 22--32. Springer, 2024

    2024

  89. [97]

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

    Ruiz, N., Li, Y., Jampani, V., Pritch, Y., Rubinstein, M., and Aberman, K. Dreambooth: Fine tuning text-to-image diffusion models for subject-driven generation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.\ 22500--22510, 2023

    2023

  90. [98]

    L., Ghasemipour, K., Gontijo Lopes, R., Karagol Ayan, B., Salimans, T., et al

    Saharia, C., Chan, W., Saxena, S., Li, L., Whang, J., Denton, E. L., Ghasemipour, K., Gontijo Lopes, R., Karagol Ayan, B., Salimans, T., et al. Photorealistic text-to-image diffusion models with deep language understanding. Advances in neural information processing systems, 35...

    2022

  91. [99]

    and Ho, J

    Salimans, T. and Ho, J. Progressive distillation for fast sampling of diffusion models. arXiv preprint arXiv:2202.00512, 2022

    2022 arXiv

  92. [100]

    and Tsaftaris, S

    Sanchez, P. and Tsaftaris, S. A. Diffusion causal models for counterfactual estimation. arXiv preprint arXiv:2202.10166, 2022

    2022 arXiv

  93. [101]

    Q., and Tsaftaris, S

    Sanchez, P., Kascenas, A., Liu, X., O’Neil, A. Q., and Tsaftaris, S. A. What is healthy? generative counterfactual diffusion for lesion localization. In MICCAI Workshop on Deep Generative Models, pp.\ 34--44. Springer, 2022 a

    2022

  94. [102]

    P., Xia, T., Watson, H

    Sanchez, P., Voisey, J. P., Xia, T., Watson, H. I., O’Neil, A. Q., and Tsaftaris, S. A. Causal machine learning for healthcare and precision medicine. Royal Society Open Science, 9 0 (8): 0 220638, 2022 b

    2022

  95. [103]

    and Geiger, A

    Sauer, A. and Geiger, A. Counterfactual generative networks. arXiv preprint arXiv:2101.06046, 2021

    2021 arXiv

  96. [104]

    Radio-opaque artefacts in digital mammography: automatic detection and analysis of downstream effects

    Schueppert, A., Glocker, B., and Roschewitz, M. Radio-opaque artefacts in digital mammography: automatic detection and analysis of downstream effects. arXiv preprint arXiv:2410.03809, 2024

    2024 arXiv

  97. [105]

    Generating interpretable counterfactual explanations by implicit minimisation of epistemic and aleatoric uncertainties

    Schut, L., Key, O., Mc Grath, R., Costabello, L., Sacaleanu, B., Gal, Y., et al. Generating interpretable counterfactual explanations by implicit minimisation of epistemic and aleatoric uncertainties. In International Conference on Artificial Intelligence and Statistics, pp.\ ...

    2021

  98. [106]

    Weakly supervised disentangled generative causal representation learning

    Shen, X., Liu, F., Dong, H., Lian, Q., Chen, Z., and Zhang, T. Weakly supervised disentangled generative causal representation learning. Journal of Machine Learning Research, 23 0 (241): 0 1--55, 2022

    2022

  99. [107]

    Promptable counterfactual diffusion model for unified brain tumor segmentation and generation with mris

    Shen, Y., He, G., and Unberath, M. Promptable counterfactual diffusion model for unified brain tumor segmentation and generation with mris. In International Workshop on Foundation Models for General Medical AI, pp.\ 81--90. Springer, 2024

    2024

  100. [108]

    and Ermon, S

    Shu, R. and Ermon, S. Bit prioritization in variational autoencoders via progressive coding. In Chaudhuri, K., Jegelka, S., Song, L., Szepesvari, C., Niu, G., and Sabato, S. (eds.), Proceedings of the 39th International Conference on Machine Learning, volume 162 of Proceedings...

    2022

  101. [109]

    Deep unsupervised learning using nonequilibrium thermodynamics

    Sohl-Dickstein, J., Weiss, E., Maheswaranathan, N., and Ganguli, S. Deep unsupervised learning using nonequilibrium thermodynamics. In Bach, F. and Blei, D. (eds.), Proceedings of the 32nd International Conference on Machine Learning, volume 37 of Proceedings of Machine Learni...

    2015

  102. [110]

    Learning structured output representation using deep conditional generative models

    Sohn, K., Lee, H., and Yan, X. Learning structured output representation using deep conditional generative models. In Cortes, C., Lawrence, N., Lee, D., Sugiyama, M., and Garnett, R. (eds.), Advances in Neural Information Processing Systems, volume 28. Curran Associates, Inc.,...

    2015

  103. [111]

    Denoising diffusion implicit models

    Song, J., Meng, C., and Ermon, S. Denoising diffusion implicit models. arXiv preprint arXiv:2010.02502, 2020 a

    2010 arXiv

  104. [112]

    Doubly abductive counterfactual inference for text-based image editing

    Song, X., Cui, J., Zhang, H., Chen, J., Hong, R., and Jiang, Y.-G. Doubly abductive counterfactual inference for text-based image editing. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.\ 9162--9171, 2024

    2024

  105. [113]

    and Ermon, S

    Song, Y. and Ermon, S. Generative modeling by estimating gradients of the data distribution. Advances in neural information processing systems, 32, 2019

    2019

  106. [114]

    P., Kumar, A., Ermon, S., and Poole, B

    Song, Y., Sohl-Dickstein, J., Kingma, D. P., Kumar, A., Ermon, S., and Poole, B. Score-based generative modeling through stochastic differential equations. arXiv preprint arXiv:2011.13456, 2020 b

    2011 arXiv

  107. [115]

    Consistency models

    Song, Y., Dhariwal, P., Chen, M., and Sutskever, I. Consistency models. arXiv preprint arXiv:2303.01469, 2023

    2023 arXiv

  108. [116]

    Disentanglement by nonlinear ica with general incompressible-flow networks (gin)

    Sorrenson, P., Rother, C., and K \"o the, U. Disentanglement by nonlinear ica with general incompressible-flow networks (gin). arXiv preprint arXiv:2001.04872, 2020

    2001 arXiv

  109. [117]

    Locinv: Localization-aware inversion for text-guided image editing

    Tang, C., Wang, K., Yang, F., and van de Weijer, J. Locinv: Localization-aware inversion for text-guided image editing. arXiv preprint arXiv:2405.01496, 2024

    2024 arXiv

  110. [118]

    What the daam: Interpreting stable diffusion using cross attention

    Tang, R., Liu, L., Pandey, A., Jiang, Z., Yang, G., Kumar, K., Stenetorp, P., Lin, J., and Ture, F. What the daam: Interpreting stable diffusion using cross attention. arXiv preprint arXiv:2210.04885, 2022

    2022 arXiv

  111. [119]

    Causal generative explainers using counterfactual inference: a case study on the morpho-mnist dataset

    Taylor-Melanson, W., Sadeghi, Z., and Matwin, S. Causal generative explainers using counterfactual inference: a case study on the morpho-mnist dataset. Pattern Analysis and Applications, 27 0 (3): 0 89, 2024

    2024

  112. [120]

    Identifying incorrect annotations in multi-label classification data

    Thyagarajan, A., Snorrason, E., Northcutt, C., and Mueller, J. Identifying incorrect annotations in multi-label classification data. arXiv preprint arXiv:2211.13895, 2022

    2022 arXiv

  113. [121]

    Diffuse, attend, and segment: Unsupervised zero-shot segmentation using stable diffusion

    Tian, J., Aggarwal, L., Colaco, A., Kira, Z., and Gonzalez-Franco, M. Diffuse, attend, and segment: Unsupervised zero-shot segmentation using stable diffusion. arXiv preprint arXiv:2308.12469, 2023

    2023 arXiv

  114. [122]

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

    Tumanyan, N., Geyer, M., Bagon, S., and Dekel, T. Plug-and-play diffusion features for text-driven image-to-image translation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.\ 1921--1930, 2023

    1921

  115. [123]

    E., Diaconu, C.-D., Markou, S., Shysheya, A., Foong, A

    Turner, R. E., Diaconu, C.-D., Markou, S., Shysheya, A., Foong, A. Y., and Mlodozeniec, B. Denoising diffusion probabilistic models in six simple steps. arXiv preprint arXiv:2402.04384, 2024

    2024 arXiv

  116. [124]

    and Kautz, J

    Vahdat, A. and Kautz, J. Nvae: A deep hierarchical variational autoencoder. Advances in neural information processing systems, 33: 0 19667--19679, 2020

    2020

  117. [125]

    and Klaise, J

    Van Looveren, A. and Klaise, J. Interpretable counterfactual explanations guided by prototypes. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases, pp.\ 650--665. Springer, 2021

    2021

  118. [126]

    u gelgen, J., Sharma, Y., Gresele, L., Brendel, W., Sch \

    Von K \"u gelgen, J., Sharma, Y., Gresele, L., Brendel, W., Sch \"o lkopf, B., Besserve, M., and Locatello, F. Self-supervised learning with data augmentations provably isolates content from style. Advances in neural information processing systems, 34: 0 16451--16467, 2021

    2021

  119. [127]

    and Vastola, J

    Wang, B. and Vastola, J. J. The hidden linear structure in score-based models and its application. arXiv preprint arXiv:2311.10892, 2023

    2023 arXiv

  120. [128]

    A., and van de Weijer, J

    Wang, K., Yang, F., Yang, S., Butt, M. A., and van de Weijer, J. Dynamic prompt learning: Addressing cross-attention leakage for text-based image editing. arXiv preprint arXiv:2309.15664, 2023 a

    2023 arXiv

  121. [129]

    Instructedit: Improving automatic masks for diffusion-based image editing with user instructions

    Wang, Q., Zhang, B., Birsak, M., and Wonka, P. Instructedit: Improving automatic masks for diffusion-based image editing with user instructions. arXiv preprint arXiv:2305.18047, 2023 b

    2023 arXiv

  122. [130]

    Concept algebra for (score-based) text-controlled generative models

    Wang, Z., Gui, L., Negrea, J., and Veitch, V. Concept algebra for (score-based) text-controlled generative models. Advances in Neural Information Processing Systems, 36, 2024

    2024

  123. [131]

    Fast diffusion-based counterfactuals for shortcut removal and generation

    Weng, N., Pegios, P., Feragen, A., Petersen, E., and Bigdeli, S. Fast diffusion-based counterfactuals for shortcut removal and generation. arXiv preprint arXiv:2312.14223, 2023

    2023 arXiv

  124. [132]

    and Paige, B

    Willetts, M. and Paige, B. I don't need u: Identifiable non-linear ica without side information. arXiv preprint arXiv:2106.05238, 2021

    2021 arXiv

  125. [133]

    Learning likelihoods with conditional normalizing flows

    Winkler, C., Worrall, D., Hoogeboom, E., and Welling, M. Learning likelihoods with conditional normalizing flows. arXiv preprint arXiv:1912.00042, 2019

    1912 arXiv

  126. [134]

    C., and Bowyer, K

    Wu, H., Bezold, G., G \"u nther, M., Boult, T., King, M. C., and Bowyer, K. W. Consistency and accuracy of celeba attribute values. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.\ 3258--3266, 2023

    2023

  127. [135]

    Counterfactual generative modeling with variational causal inference

    Wu, Y., McConnell, L., and Iriondo, C. Counterfactual generative modeling with variational causal inference. arXiv preprint arXiv:2410.12730, 2024

    2024 arXiv

  128. [136]

    M., Pan, Y., and Bareinboim, E

    Xia, K. M., Pan, Y., and Bareinboim, E. Neural causal models for counterfactual identification and estimation. In The Eleventh International Conference on Learning Representations, 2023

    2023

  129. [137]

    Xia, T., Roschewitz, M., Ribeiro, F. D. S., Jones, C., and Glocker, B. Mitigating attribute amplification in counterfactual image generation. arXiv preprint arXiv:2403.09422, 2024

    2024 arXiv

  130. [138]

    Counterfactual generation with identifiability guarantees

    Yan, H., Kong, L., Gui, L., Chi, Y., Xing, E., He, Y., and Zhang, K. Counterfactual generation with identifiability guarantees. Advances in Neural Information Processing Systems, 36: 0 56256--56277, 2023

    2023

  131. [139]

    Paint by example: Exemplar-based image editing with diffusion models

    Yang, B., Gu, S., Zhang, B., Zhang, T., Chen, X., Sun, X., Chen, D., and Wen, F. Paint by example: Exemplar-based image editing with diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.\ 18381--18391, 2023 a

    2023

  132. [140]

    A., van de Weijer, J., et al

    Yang, F., Yang, S., Butt, M. A., van de Weijer, J., et al. Dynamic prompt learning: Addressing cross-attention leakage for text-based image editing. Advances in Neural Information Processing Systems, 36, 2024

    2024

  133. [141]

    Causalvae: Disentangled representation learning via neural structural causal models

    Yang, M., Liu, F., Chen, Z., Shen, X., Hao, J., and Wang, J. Causalvae: Disentangled representation learning via neural structural causal models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.\ 9593--9602, 2021

    2021

  134. [142]

    Disdiff: Unsupervised disentanglement of diffusion probabilistic models

    Yang, T., Wang, Y., Lv, Y., and Zheng, N. Disdiff: Unsupervised disentanglement of diffusion probabilistic models. arXiv preprint arXiv:2301.13721, 2023 b

    2023 arXiv

  135. [143]

    Pearl causal hierarchy on image data: Intricacies & challenges

    Ze c evi \'c , M., Willig, M., Singh Dhami, D., and Kersting, K. Pearl causal hierarchy on image data: Intricacies & challenges. arXiv e-prints, pp.\ arXiv--2212, 2022

    2022

  136. [144]

    A., Shechtman, E., and Wang, O

    Zhang, R., Isola, P., Efros, A. A., Shechtman, E., and Wang, O. The unreasonable effectiveness of deep features as a perceptual metric. In CVPR, 2018

    2018

  137. [145]

    Unsupervised representation learning from pre-trained diffusion probabilistic models

    Zhang, Z., Zhao, Z., and Lin, Z. Unsupervised representation learning from pre-trained diffusion probabilistic models. Advances in neural information processing systems, 35: 0 22117--22130, 2022

    2022

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.