Pith. sign in

REVIEW 4 major objections 5 minor 36 references

VSC: Visual Search Compositional Text-to-Image Diffusion Model

T0 review · 4 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read The paper claims that generating a reference image for each attribute-object pair separately, averaging those images into visual prototypes, and fusing the prototypes into text embeddings improves attribute-object binding in text-to-image…

desk verdict A sensible compositional generation method with consistent benchmark gains, but the training-filter/eval-metric overlap ambiguity in §3.3 must be resolved before the headline claims are clean. read the letter →

arxiv 2505.01104 v1 pith:NGZWAFFY submitted 2025-05-02 cs.CV

Do Huu Dat , Nam Hyeonu , Po-Yuan Mao , Tae-Hyun Oh This is my paper · ORCID
classification cs.CV
keywords text-to-imagegenerationattribute-objectbindingcompositionalvisualprototypecross-attentionlocalizationsyntheticdatacurationT2ICompBenchBLIP-VQA
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-to-image diffusion models often misbind attributes: given a prompt like "a red car and a yellow bicycle," they may swap or blend the colors. VSC attacks this by decomposing the prompt into single attribute-object pairs, generating a reference image for each pair with the same frozen diffusion model, and averaging the image encodings into a visual prototype that is injected into the text-embedding stream via a small trained MLP. A segmentation-based localization loss aligns the cross-attention maps of attribute and object tokens with masks, correcting a known source of binding errors. The paper reports consistent gains over prior compositional methods on the T2I CompBench benchmark for color, texture, and shape, on three Stable Diffusion backbones, and shows the advantage widening as the number of binding pairs grows. A human evaluation also prefers VSC's images when objects are judged in isolation.

What carries the argument

The load-bearing object is the pairwise visual prototype: for each binding pair $[a_n, o_n]$, the method generates $m$ reference images with the pre-trained model, encodes each with an image encoder $\phi$, and takes the mean $\mathbf{e}_j = \frac{1}{m}\sum_k \phi(r_j^k)$ as that pair's prototype. At the token positions belonging to an attribute or object, the text embedding $\mathbf{c}_i$ is replaced by $\mathrm{MLP}([\mathbf{c}_i, \mathbf{e}_j])$, so the frozen diffusion model conditions on a visual hint of the intended pair. The second mechanism is the segmentation-based localization loss, which drives the cross-attention maps of both the attribute token and the object token toward the object's segmentation mask, preventing one pair's attention from leaking into another's pixels. Together they convert the model's reliable single-pair generation into multi-pair binding without changing the generation backbone.

What would settle it

Compare VSC against SynGen on a held-out set of novel attribute-object pairs where human annotators, not BLIP-VQA, judge whether each object carries its stated attribute; if VSC's margin shrinks or disappears, the reported advantage is partly an artifact of metric-matched training. A quicker check: retrain the MLP using a different filtering score (or no filtering) and see whether the CompBench numbers drop.

Watch

Extended reading notes

Core claim

The central claim is that a frozen pre-trained diffusion model can bind multiple attribute-object pairs correctly if it is shown what each pair looks like on its own. Taking the prompt's pairs, the method generates several images per pair, encodes them with an image encoder, and averages the encodings into a visual prototype embedding for that pair. Those prototypes are fused into the text embeddings at the attribute and object token positions through a trained MLP, and a localization loss pushes the cross-attention maps of each attribute and its object onto the object's segmentation mask. On T2I CompBench, VSC reports BLIP-VQA scores of 0.66/0.61/0.47 on Stable Diffusion 1.4, 0.74/0.64/0.53 on 2.1, and 0.85/0.79/0.63 on 3.5 for color/texture/shape, each above the best prior method on the same backbone, and it outperforms SynGen by an increasing margin at 3, 4, and 5 binding pairs.

Load-bearing premise

The dataset curation selects images by a metric that appears to be the same BLIP-VQA score used for evaluation, so the reported advantage could partly reflect optimization toward that scorer rather than genuinely better attribute binding.

Editorial extensions

If this is right

  • On Stable Diffusion 3.5, VSC raises the harmonic mean on T2I CompBench to 0.727 from SynGen's 0.703, with the largest single gain in shape (0.63 vs. 0.59).
  • The margin over SynGen grows with prompt complexity: at five binding pairs, VSC scores 0.246 on SD 3.5 versus SynGen's 0.1815, indicating that the visual prototypes buffer against attention-map degradation.
  • The recipe is backbone-agnostic: the same training procedure improves Stable Diffusion 1.4, 2.1, and 3.5, so stronger base models directly yield better binding.
  • Training only on color pairs transfers to texture and shape (e.g., SD 3.5 color-only training reaches 0.74 texture vs. 0.67 baseline), suggesting the MLP learns a general fusion operation rather than color-specific heuristics.
  • Human raters chose VSC-generated objects as highest-quality 30.01% of the time versus 25.18% for SynGen, even when judging crops detached from binding correctness.

Reading between the lines

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

  • The method's reliance on the base model's single-pair skill suggests a testable recipe: if the reference generator is upgraded (for example, a stronger text-to-image model), VSC's multi-pair accuracy should rise without retraining the fusion MLP.
  • Because the synthetic training set is filtered by the same BLIP-VQA metric used for evaluation, part of the reported advantage may reflect optimization toward that scorer; an independent human binding test or a different visual-question-answering metric would reveal how much of the gain is real.
  • The localization loss requires segmentation masks at training time, but inference needs none; one could try replacing the masks with attention-derived pseudo-masks during training, which would make the approach applicable to models and domains where instance segmentation is unavailable.
  • The scaling curve (improvement saturating near 30k images) hints that the bottleneck is reference-image quality rather than dataset size; sampling more diverse, higher-fidelity reference images per pair may be a cheaper path to further gains.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The manuscript proposes VSC (Visual Search Composition), a method for improving attribute-object binding in text-to-image diffusion models. Given a prompt with multiple attribute-object pairs, VSC generates one image per pair with a frozen pre-trained diffusion model, encodes these reference images with a CLIP image encoder, averages the embeddings per pair to form visual prototypes, and fuses these prototypes into the text embeddings at the corresponding token positions through a lightweight MLP. The MLP and the last few image-encoder layers are fine-tuned on a synthetically created dataset, and a cross-attention localization loss derived from FastComposer is added to align attribute/object attention maps with segmentation masks. Experiments on T2I-CompBench for color, texture, and shape report state-of-the-art BLIP-VQA scores on Stable Diffusion 1.4, 2.1, and 3.5, plus a small human study, an analysis of scaling the number of binding pairs, dataset-size scaling, and transferability from color to other attribute categories.

Significance. If the reported results are taken at face value, the paper makes a useful empirical contribution: it adapts subject-driven image generation to attribute binding without user-provided layouts, requires training only an MLP and a few image-encoder layers, and shows consistent improvements across three Stable Diffusion backbones. The synthetic-data pipeline and the localization-loss design are simple and reasonably motivated by the single-pair generation ability of pre-trained models. The paper also provides a scaling-law analysis and a transferability study that go beyond a single benchmark table. However, the central claim depends on the BLIP-VQA benchmark numbers, and the dataset-creation step contains a potentially circular use of that same metric; until that ambiguity is resolved and basic statistical reporting is added, the SOTA claim is not fully supported.

major comments (4)
  1. [Section 3.3 (Dataset Creation)] The filtering sentence "Finally, we select 45 images with the best scores for each prompt" is ambiguous because the preceding paragraph describes both OpenCLIP alignment scores and BLIP-VQA scores. If BLIP-VQA is the score used for filtering, then the training reference images are selected by the exact metric used for evaluation in Tables 1 and 2; since the visual prototypes are averages of embeddings from these selected images, the model could learn to produce images that score well on BLIP-VQA without genuinely improving attribute-object binding, inflating the gains over SynGen and other baselines. This is load-bearing for the central SOTA claim. Please state unambiguously which score is used for filtering, and report an ablation where reference images are selected by OpenCLIP only or randomly, evaluated on the same benchmark.
  2. [Section 4.2, Tables 1 and 2] No error bars, seeds, or confidence intervals are reported for any numerical result. Several differences over strong baselines are small (e.g., SD1.4 color 0.66 vs. SynGen 0.63, shape 0.47 vs. 0.46; SD3.5 color 0.85 vs. 0.82), and the scaling-pairs results in Table 2 are single numbers per configuration. Without variance estimates or the number of prompts per cell, these differences may not be statistically significant. Please report at least three random seeds with means and confidence intervals for every cell, and specify how many prompts are used in Table 2.
  3. [Section 4.2.1 (Human Eval)] The human evaluation is described as using ten raters with majority voting and an "equally good" / "equally bad" option, but the manuscript does not report the number of prompts, the number of pairwise comparisons, or any measure of rater agreement or statistical confidence (e.g., confidence intervals or Fleiss' kappa). A 30.01% vs. 25.18% majority-vote difference with no variance does not substantiate the abstract's claim of "better image quality, evaluated by humans." Please provide the full protocol, the number of judgments, and appropriate statistics.
  4. [Section 6.1, Table 5] The claim that pre-trained diffusion models are already "superior" at single attribute-object pair generation is supported only by BLIP-VQA scores for SD2.1, with no error bars or other backbones. This claim is used to justify using generated images as reference images for all three backbone models, so it should be verified for SD1.4 and SD3.5 as well, ideally with multiple seeds and at least one additional metric.
minor comments (5)
  1. [Abstract / Introduction] There is a typo in the Abstract and Introduction: "Our approaches" should be "Our approach" (singular), and the abstract's claim of "better image quality, evaluated by humans" is worded more strongly than the human study supports.
  2. [Equation (6)] The localization loss is described as a "balanced L1 loss," but the expression is a sum of mean differences, not an L1 norm; please clarify the terminology and the exact objective being minimized.
  3. [Figure 5] The figure legend mixes attribute names (Color, Shape, Texture) and backbone names (SD2.1, SD3.5) without a clear key, and the y-axis label "Accuracy" does not specify whether the metric is BLIP-VQA or something else; please make the legend and axis explicit.
  4. [Section 5.3 / Figure 6] The inference-time comparison reports a single runtime curve without error bars or a description of how many runs were averaged; please specify the hardware, number of repetitions, and variance.
  5. [References] The implementation details cite Stable Diffusion 1.4, 2.1, and 3.5 via references [8] and [25], but reference [25] is the original latent diffusion paper; please cite the appropriate Stable Diffusion model cards or releases for each version.

Circularity Check

0 steps flagged · score 0.0 of 10

No demonstrated circularity: the only candidate is an unstated dataset-filter metric in §3.3, which is a verification gap rather than a shown reduction.

full rationale

The paper's method is empirical rather than derivational, so there is no chain of equations that could reduce to its inputs. VSC decomposes prompts, generates single-pair reference images with a pretrained diffusion model, averages their image embeddings into prototypes, fuses them with text embeddings via a trainable MLP, and trains with FastComposer's localization loss. The central claim is checked on the external T2I-CompBench benchmark against published baselines (SynGen, Attend-and-Excite, etc.), and BLIP-VQA is an external evaluator, not a term in the training loss. The single plausible circularity candidate is Section 3.3's sentence 'Finally, we select 45 images with the best scores for each prompt.' The surrounding text mentions both an OpenCLIP alignment score, used for mask assignment, and the BLIP-VQA evaluation metric, but never states which score filters the training images. Because the filtering metric is unstated, the text does not exhibit the specific reduction 'train on BLIP-VQA-selected data, then evaluate with BLIP-VQA'; asserting that reduction would require speculation about author intent. If the filter had been BLIP-VQA, the reported gains over SynGen in Tables 1-2 would be partly metric-driven, but that is a correctness/leakage concern, not a demonstrated self-definitional or fitted-prediction circularity under the evidentiary rules. No load-bearing self-citations, author-imported uniqueness theorems, or ansatz-smuggling citations appear in the manuscript. Hence the circularity score is 0.

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

The method relies on existing components: pretrained diffusion models, CLIP image encoders, Mask2Former segmentation, and the BLIP-VQA metric. No new entities are introduced. The primary unevidenced assumptions are the reliability of the model's own single-pair generations as references and the use of the evaluation metric in data curation.

free parameters (4)
  • lambda (localization loss weight) = not reported
    Appears in the objective L = Lnoise + lambda * Lloc in Section 3.2, but no value is given.
  • m (number of reference images per pair) = not reported
    Used in Eq. 4 as the number of generated reference images per pair; no value is stated for training or inference.
  • Dataset filtering threshold = 45 images per prompt
    Section 3.3 selects the 45 highest-scoring images per prompt; this choice shapes the training distribution.
  • Fine-tuned image encoder layers = 2 (ViT-L/14) or 3 (ViT-H/14)
    From Fig. 7, the last 2 or 3 layers are trainable; no ablation of this choice is reported.
assumptions (4)
  • domain assumption The pretrained diffusion model generates sufficiently accurate single-pair images to serve as references.
    Central to Eq. 3-4; supported by Table 5 only for SD2.1, with no variance reported.
  • domain assumption BLIP-VQA is a valid proxy for attribute-binding quality.
    Adopted from Huang et al. [13]; used both to filter training data and to evaluate.
  • domain assumption Mask2Former segmentation masks on synthetic images are accurate enough to supervise the localization loss.
    Lloc in Eq. 6 uses these masks; no accuracy analysis on the synthetic dataset is provided.
  • domain assumption CLIP image embeddings averaged over reference images preserve the attribute-object relationship.
    The visual prototype e_j in Eq. 4 is a mean of CLIP embeddings; the paper does not demonstrate that averaging does not dilute the attribute.

how reviews work

0 comments
Cite this review

Pith. "Pith review of VSC: Visual Search Compositional Text-to-Image Diffusion Model." pith.science (2026). https://pith.science/paper/NGZWAFFY

@misc{pith2026250501104,
  author       = {Pith},
  title        = {Pith review of: VSC: Visual Search Compositional Text-to-Image Diffusion Model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NGZWAFFY}},
  note         = {Machine review of arXiv:2505.01104}
}
read the original abstract

Text-to-image diffusion models have shown impressive capabilities in generating realistic visuals from natural-language prompts, yet they often struggle with accurately binding attributes to corresponding objects, especially in prompts containing multiple attribute-object pairs. This challenge primarily arises from the limitations of commonly used text encoders, such as CLIP, which can fail to encode complex linguistic relationships and modifiers effectively. Existing approaches have attempted to mitigate these issues through attention map control during inference and the use of layout information or fine-tuning during training, yet they face performance drops with increased prompt complexity. In this work, we introduce a novel compositional generation method that leverages pairwise image embeddings to improve attribute-object binding. Our approach decomposes complex prompts into sub-prompts, generates corresponding images, and computes visual prototypes that fuse with text embeddings to enhance representation. By applying segmentation-based localization training, we address cross-attention misalignment, achieving improved accuracy in binding multiple attributes to objects. Our approaches outperform existing compositional text-to-image diffusion models on the benchmark T2I CompBench, achieving better image quality, evaluated by humans, and emerging robustness under scaling number of binding pairs in the prompt.

Figures

Figures reproduced from arXiv: 2505.01104 by the authors.

Figure 1
Figure 1. Left: Original pipeline and generated image of Diffusion Model. Right: VSC individually “search” for visual information of each binding pair by generating images for each separate pair (“red car” and “yellow bicycle”). Finally, VSC uses them as references to generate the correct compositional image. Abstract Text-to-image diffusion models have shown impressive capa￾bilities in generating realistic visuals from natur… view at source ↗
Figure 2
Figure 2. The training pipeline of VSC. Given the input text prompt “A blue apple and a green backpack”, VSC first generates images of each binding pair (“blue apple” and “green backpack”), then uses an image encoder to extract visual prototype features, which is later used to augment the text embeddings via a trainable MLP module. Finally, a frozen pre-trained diffusion model generates the image with the augmented text embed… view at source ↗
Figure 3
Figure 3. Generation with the increasing number of binding pairs. We show the generated images when the prompt includes more pairs. We observe that the generated image consistently reflects the additional composition [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Qualitative result of compositionally. Compared to the baselines, our method can generate the image with better composition. Specifically, for the last row, Stable Diffusion 3.5 and VSC with 2.1 and 3.5 reflect the prompt better than others, such as silver refrigerator…
Figure 5
Figure 5. Figure 5: Scaling dataset. The accuracy of models on T2I￾CompBench while training on different dataset sizes, highlighting the positive impact of dataset scaling on model performance. We trained our model on six different subsets, each con￾taining 5k, 10k, 20k, 30k, 60k, and 90k…
Figure 6
Figure 6. Figure 6: Inference cost. We show the inference time of VSC, SynGen, A&E, and the baseline according to the number of bind￾ing pairs in the prompt. Although VSC is the slowest in 2-pairs generation, VSC efficiently scales to be faster than SynGen and A&E after the number of pair…
Figure 8
Figure 8. Figure 8: Visualization of cross-attention maps. In VSC, the maps are more fine-grained, and the pairs’ maps are well-aligned. to revert the image back to timestep 20, feed-forward to the UNet backbone, and visualize the cross-attention maps of both models at that timestep [PIT…
Figure 7
Figure 7. Figure 7: Localization cross-attention Loss (LLoc). We depict the importance of fine-tuning the last few layers of the image encoder, as it help the whole pipeline to descent the LLoc [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 19 canonical work pages

  1. [1]

    A-star: Test-time attention segregation and retention for text- to-image synthesis

    Aishwarya Agarwal, Srikrishna Karanam, KJ Joseph, Apoorv Saxena, Koustava Goswami, and Balaji Vasan Srinivasan. A-star: Test-time attention segregation and retention for text- to-image synthesis. In Proceedings of the IEEE/CVF Inter- national Conference on Computer Vision, pages 2283–2293,

    work page

  2. [2]

    Spatext: Spatio-textual representation for con- trollable image generation

    Omri Avrahami, Thomas Hayes, Oran Gafni, Sonal Gupta, Yaniv Taigman, Devi Parikh, Dani Lischinski, Ohad Fried, and Xi Yin. Spatext: Spatio-textual representation for con- trollable image generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 18370–18380, 2023. 3

    work page 2023

  3. [3]

    Understanding the limits of vision language mod- els through the lens of the binding problem

    Declan Campbell, Sunayana Rane, Tyler Giallanza, Nicol`o De Sabbata, Kia Ghods, Amogh Joshi, Alexander Ku, Steven M Frankland, Thomas L Griffiths, Jonathan D Co- hen, et al. Understanding the limits of vision language mod- els through the lens of the binding problem. arXiv preprint arXiv:2411.00238, 2024. 2

    arXiv 2024

  4. [4]

    Attend-and-excite: Attention-based semantic guid- ance for text-to-image diffusion models

    Hila Chefer, Yuval Alaluf, Yael Vinker, Lior Wolf, and Daniel Cohen-Or. Attend-and-excite: Attention-based semantic guid- ance for text-to-image diffusion models. ACM Transactions on Graphics (TOG), 42(4):1–10, 2023. 5

    work page 2023

  5. [5]

    Schwing, Alexan- der Kirillov, and Rohit Girdhar

    Bowen Cheng, Ishan Misra, Alexander G. Schwing, Alexan- der Kirillov, and Rohit Girdhar. Masked-attention mask trans- former for universal image segmentation. 2022. 4

    work page 2022

  6. [6]

    Aspects of the Theory of Syntax

    Noam Chomsky. Aspects of the Theory of Syntax. Number 11. MIT press, 2014. 1

    work page 2014

  7. [7]

    Be yourself: Bounded attention for multi-subject text-to-image generation

    Omer Dahary, Or Patashnik, Kfir Aberman, and Daniel Cohen-Or. Be yourself: Bounded attention for multi-subject text-to-image generation. arXiv preprint arXiv:2403.16990, 2(5), 2024. 2

    arXiv 2024

  8. [8]

    Scaling rectified flow trans- formers for high-resolution image synthesis

    Patrick Esser, Sumith Kulal, Andreas Blattmann, Rahim En- tezari, Jonas M¨uller, Harry Saini, Yam Levi, Dominik Lorenz, Axel Sauer, Frederic Boesel, et al. Scaling rectified flow trans- formers for high-resolution image synthesis. In Forty-first International Conference on Machine Learning, 2024. 4

    work page 2024

Show all 36 references
  1. [9]

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

    Weixi Feng, Xuehai He, Tsu-Jui Fu, Varun Jampani, Arjun Akula, Pradyumna Narayana, Sugato Basu, Xin Eric Wang, and William Yang Wang. Training-free structured diffusion guidance for compositional text-to-image synthesis. arXiv preprint arXiv:2212.05032, 2022. 2, 4

    2022 arXiv

  2. [10]

    Layoutgpt: Compositional visual plan- ning and generation with large language models

    Weixi Feng, Wanrong Zhu, Tsu-jui Fu, Varun Jampani, Ar- jun Akula, Xuehai He, Sugato Basu, Xin Eric Wang, and William Yang Wang. Layoutgpt: Compositional visual plan- ning and generation with large language models. Advances in Neural Information Processing Systems, 36, 2024. 2

    2024

  3. [11]

    An image is worth one word: Personalizing text-to-image genera- tion using textual inversion.arXiv preprint arXiv:2208.01618,

    Rinon Gal, Yuval Alaluf, Yuval Atzmon, Or Patashnik, Amit H Bermano, Gal Chechik, and Daniel Cohen-Or. An image is worth one word: Personalizing text-to-image genera- tion using textual inversion.arXiv preprint arXiv:2208.01618,

    arXiv

  4. [12]

    Encoder-based domain tuning for fast personalization of text-to-image models

    Rinon Gal, Moab Arar, Yuval Atzmon, Amit H Bermano, Gal Chechik, and Daniel Cohen-Or. Encoder-based domain tuning for fast personalization of text-to-image models. ACM Transactions on Graphics (TOG), 42(4):1–13, 2023. 3

    2023

  5. [13]

    T2i-compbench: A comprehensive benchmark for open- world compositional text-to-image generation

    Kaiyi Huang, Kaiyue Sun, Enze Xie, Zhenguo Li, and Xihui Liu. T2i-compbench: A comprehensive benchmark for open- world compositional text-to-image generation. Advances in Neural Information Processing Systems , 36:78723–78747,

  6. [14]

    Openclip, 2021

    Gabriel Ilharco, Mitchell Wortsman, Ross Wightman, Cade Gordon, Nicholas Carlini, Rohan Taori, Achal Dave, Vaishaal Shankar, Hongseok Namkoong, John Miller, Hannaneh Ha- jishirzi, Ali Farhadi, and Ludwig Schmidt. Openclip, 2021. If you use this software, please cite it as below. 4

    2021

  7. [15]

    X&fuse: Fusing visual information in text-to-image generation

    Yuval Kirstain, Omer Levy, and Adam Polyak. X&fuse: Fusing visual information in text-to-image generation. arXiv preprint arXiv:2303.01000, 2023. 3

    2023 arXiv

  8. [16]

    Multi-concept customization of text- to-image diffusion

    Nupur Kumari, Bingliang Zhang, Richard Zhang, Eli Shecht- man, and Jun-Yan Zhu. Multi-concept customization of text- to-image diffusion. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 1931–1941, 2023. 2, 3

    1931

  9. [17]

    Does clip bind concepts? probing compositionality in large image models

    Martha Lewis, Nihal V Nayak, Peilin Yu, Qinan Yu, Jack Merullo, Stephen H Bach, and Ellie Pavlick. Does clip bind concepts? probing compositionality in large image models. arXiv preprint arXiv:2212.10537, 2022. 2

    2022 arXiv

  10. [18]

    Compositional visual generation with composable diffusion models

    Nan Liu, Shuang Li, Yilun Du, Antonio Torralba, and Joshua B Tenenbaum. Compositional visual generation with composable diffusion models. In European Conference on Computer Vision, pages 423–439. Springer, 2022. 2, 4

    2022

  11. [19]

    Conform: Contrast is all you need for high- fidelity text-to-image diffusion models

    Tuna Han Salih Meral, Enis Simsar, Federico Tombari, and Pinar Yanardag. Conform: Contrast is all you need for high- fidelity text-to-image diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9005–9014, 2024. 2, 5

    2024

  12. [20]

    Null-text inversion for editing real im- ages using guided diffusion models

    Ron Mokady, Amir Hertz, Kfir Aberman, Yael Pritch, and Daniel Cohen-Or. Null-text inversion for editing real im- ages using guided diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 6038–6047, 2023. 3

    2023

  13. [21]

    Compositional text-to-image generation with dense blob representations

    Weili Nie, Sifei Liu, Morteza Mardani, Chao Liu, Benjamin Eckart, and Arash Vahdat. Compositional text-to-image generation with dense blob representations. arXiv preprint arXiv:2405.08246, 2024. 2

    2024 arXiv

  14. [22]

    Compositional abilities emerge multiplicatively: Ex- ploring diffusion models on a synthetic task

    Maya Okawa, Ekdeep S Lubana, Robert Dick, and Hidenori Tanaka. Compositional abilities emerge multiplicatively: Ex- ploring diffusion models on a synthetic task. Advances in Neural Information Processing Systems, 36, 2024. 1

    2024

  15. [23]

    Learning transferable visual models from natural language supervi- sion

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

    2021

  16. [24]

    Linguistic binding in dif- fusion models: Enhancing attribute correspondence through attention map alignment

    Royi Rassin, Eran Hirsch, Daniel Glickman, Shauli Ravfogel, Yoav Goldberg, and Gal Chechik. Linguistic binding in dif- fusion models: Enhancing attribute correspondence through attention map alignment. Advances in Neural Information Processing Systems, 36, 2024. 2, 4, 5

    2024

  17. [25]

    High-resolution image 9 synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj ¨orn Ommer. High-resolution image 9 synthesis with latent diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022. 4

    2022

  18. [26]

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

    Nataniel Ruiz, Yuanzhen Li, Varun Jampani, Yael Pritch, Michael Rubinstein, and Kfir Aberman. Dreambooth: Fine tuning text-to-image diffusion models for subject-driven gen- eration. In Proceedings of the IEEE/CVF conference on com- puter vision and pattern recognition , pages ...

  19. [27]

    Collage diffusion

    Vishnu Sarukkai, Linden Li, Arden Ma, Christopher R´e, and Kayvon Fatahalian. Collage diffusion. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 4208–4217, 2024. 3

    2024

  20. [28]

    In- stantbooth: Personalized text-to-image generation without test-time finetuning

    Jing Shi, Wei Xiong, Zhe Lin, and Hyun Joon Jung. In- stantbooth: Personalized text-to-image generation without test-time finetuning. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 8543–8552, 2024. 3

    2024

  21. [29]

    Diffusion lens: Interpreting text encoders in text-to-image pipelines

    Michael Toker, Hadas Orgad, Mor Ventura, Dana Arad, and Yonatan Belinkov. Diffusion lens: Interpreting text encoders in text-to-image pipelines. arXiv preprint arXiv:2403.05846,

    arXiv

  22. [30]

    Eyes wide shut? exploring the visual shortcomings of multimodal llms

    Shengbang Tong, Zhuang Liu, Yuexiang Zhai, Yi Ma, Yann LeCun, and Saining Xie. Eyes wide shut? exploring the visual shortcomings of multimodal llms. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9568–9578, 2024. 2

    2024

  23. [31]

    A feature-integration theory of attention

    Anne M Treisman and Garry Gelade. A feature-integration theory of attention. Cognitive psychology, 12(1):97–136,

  24. [32]

    Compositional text-to-image synthesis with attention map control of diffusion models

    Ruichen Wang, Zekang Chen, Chen Chen, Jian Ma, Haonan Lu, and Xiaodong Lin. Compositional text-to-image synthesis with attention map control of diffusion models. In Proceed- ings of the AAAI Conference on Artificial Intelligence, pages 5544–5552, 2024. 2

    2024

  25. [33]

    Elite: Encoding visual concepts into textual embeddings for customized text-to-image generation

    Yuxiang Wei, Yabo Zhang, Zhilong Ji, Jinfeng Bai, Lei Zhang, and Wangmeng Zuo. Elite: Encoding visual concepts into textual embeddings for customized text-to-image generation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 15943–15953, 2023. 3

    2023

  26. [34]

    Fastcomposer: Tuning-free multi- subject image generation with localized attention

    Guangxuan Xiao, Tianwei Yin, William T Freeman, Fr ´edo Durand, and Song Han. Fastcomposer: Tuning-free multi- subject image generation with localized attention. Interna- tional Journal of Computer Vision, pages 1–20, 2024. 2, 3, 4

    2024

  27. [35]

    Understanding and mitigating compositional issues in text-to-image generative models

    Arman Zarei, Keivan Rezaei, Samyadeep Basu, Mehrdad Saberi, Mazda Moayeri, Priyatham Kattakinda, and So- heil Feizi. Understanding and mitigating compositional issues in text-to-image generative models. arXiv preprint arXiv:2406.07844, 2024. 2, 5, 8

    2024 arXiv

  28. [36]

    Layoutdiffusion: Controllable diffusion model for layout-to-image generation

    Guangcong Zheng, Xianpan Zhou, Xuewei Li, Zhongang Qi, Ying Shan, and Xi Li. Layoutdiffusion: Controllable diffusion model for layout-to-image generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 22490–22499, 2023. 2 10

    2023

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.