Pith. sign in

REVIEW 5 major objections 7 minor 60 references

Learning Graph Representation of Agent Diffusers

T0 review · 5 major / 7 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read LGR-AD claims that coordinating diffusion models as interacting agents on a graph, pruned to a maximum spanning tree, improves text-to-image generation on MSCOCO, CUB, LN-COCO, and MM CelebA-HQ.

desk verdict LGR-AD's graph-ensemble framing is plausible, but the missing decoding step and sign error in the loss sink the reported results. read the letter →

arxiv 2505.06761 v2 pith:ZF5K5AMI submitted 2025-05-10 cs.LG cs.MA

Youcef Djenouri , Nassim Belmecheri , Tomasz Michalak , Jan Dubiński , Ahmed Nabil Belbachir , Anis Yazidi This is my paper · ORCID
classification cs.LGcs.MA
keywords multi-agentsystemsdiffusionmodelstext-to-imagegenerationgraphneuralnetworksmaximumspanningtreemodelensemblingdiversity-accuracytrade-offrepresentationlearning
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 show that text-to-image generation improves when individual diffusion models are treated as agents in a multi-agent system, connected by a graph and coordinated by a graph neural network, rather than relying on any single model or a naive ensemble. The proposed LGR-AD builds a graph whose nodes are trained diffusers and whose edges are weighted by how similar the models' specifications or outputs are, then extracts a maximum spanning tree to keep only the strongest relationships. A graph convolutional network learns embeddings from this tree and a fully connected head produces the final image. On MSCOCO, CUB, LN-COCO, and MM CelebA-HQ, the paper reports lower FID, higher Inception Score, and higher CLIP score than standalone diffusion models and the MagicFusion ensemble baseline. If true, this would give a principled way to combine off-the-shelf generative models without running every expert model at full cost.

What carries the argument

The central machinery is the graph-of-models representation together with the top-$k$ maximum spanning tree (MST) and the GCNN meta-model. The MST is the acyclic subgraph that maximizes total edge weight, so it keeps only the strongest model relationships and removes redundant cycles; in the experiments a single MST is sampled ($k=1$). The GCNN layer update $H^{(l+1)} = \sigma(\tilde{M} H^{(l)} W^{(l)})$ propagates information along those selected edges, and the Laplacian loss $L_{\text{laplace}} = \frac{1}{2}\sum_{i,j} A^{\text{MST}}_{ij}\|h_i - h_j\|^2$ forces strongly connected diffusers to have similar embeddings. These pieces are trained with $L(x) = C(x) + \lambda D(x) + \gamma L_{\text{laplace}}$, the composite loss the paper says balances accuracy, diversity, and graph structure.

What would settle it

Train LGR-AD with the KL coefficient $\lambda$ set to $0$ and with $\lambda > 0$, keeping everything else fixed, and measure the diversity of generated images (e.g., average pairwise embedding distance). If the printed positive sign drives diversity, diversity should increase with $\lambda$, whereas the loss-minimization direction predicts it decreases; independently, remove the Laplacian loss ($\gamma=0$) and replace the MST with the full graph to see whether the graph structure or the extra loss terms explain the reported gains.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central claim is that ensemble coordination can be cast as graph representation learning over the space of expert diffusers. Each diffuser is a node; two connectivity functions, one over model architectures (CCF) and one over model outputs (PCF), assign edge weights; a maximum spanning tree selects the most informative relationships; and a GCNN trained with a hybrid loss (cross-entropy for accuracy, KL divergence for diversity, and graph Laplacian for structural smoothness) learns node embeddings that guide final image synthesis. The authors assert that this pipeline consistently beats the individual models and the MagicFusion ensemble on all four benchmarks and all three metrics, and that the full four-model graph with hybrid connectivity is better than any subgraph.

Load-bearing premise

The accuracy-diversity balance rests on the claim that adding the KL-divergence term $D(x)$ with a positive coefficient makes the ensemble more diverse; in the loss as written, minimizing $L$ also minimizes $D$, so this only works if the intended sign is negative or the described mechanism is different.

Editorial extensions

If this is right

  • If LGR-AD works as claimed, existing pre-trained diffusion models can be combined without retraining them, because only a small GCNN meta-model needs to be trained.
  • The MST-based pruning means only the most informative model relationships are used, so the approach scales more cheaply than executing every expert diffuser in the pool.
  • The hybrid CCF+PCF connectivity should generally be preferred over either connectivity function alone, since it captures both architectural and output-level relationships between agents.
  • Because the reported gains hold across a fine-grained bird dataset, a noisy COCO variant, and a face dataset, the method is claimed to be dataset-agnostic rather than tuned to one domain.

Reading between the lines

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

  • A direct test the paper leaves implicit is to flip the sign of the KL term in Eq. (10): the text says diversity is promoted by maximizing KL, but the printed loss minimizes $D(x)$, so the mechanism would only work as described if the intended loss is $L(x) = C(x) - \lambda D(x) + \gamma L_{\text{laplace}}$.
  • The MST is only one possible graph sparsifier; comparing it against k-nearest-neighbor graphs or attention-weighted complete graphs on the same node features would show whether the spanning-tree structure itself, rather than the edge weights, drives the reported gains.
  • Since the graph is built from model outputs and specifications, the same multi-agent representation could coordinate heterogeneous generative models such as GANs or auto-regressive transformers, making LGR-AD a general model-ensembling scheme rather than a diffusion-specific one.
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

5 major / 7 minor

Summary. The paper proposes LGR-AD, a multi-agent framework in which each diffusion model is an agent and a graph is built over agents using either characteristic (architecture) or performance (output) connectivity functions. A maximum spanning tree is extracted, a GCNN is trained with a composite loss of cross-entropy, KL divergence, and a Laplacian term, and the resulting embeddings are claimed to be decoded into the final text-to-image output. The authors report state-of-the-art FID, IS, and CLIP scores on MSCOCO, CUB, LN-COCO, and MM CelebA-HQ, plus an ablation on the number of agents and connectivity functions. The central claim is that LGR-AD consistently outperforms individual diffusion models and the MagicFusion ensemble baseline.

Significance. If the method worked as described, the idea of selecting and coordinating a subset of expert diffusion models through a graph meta-model would be a useful contribution to multi-agent and ensemble-diffusion research. The paper also makes an effort to ablate design choices, and it provides a public code link. However, the present manuscript does not support the central claim: the loss objective is internally inconsistent, the graph connectivity functions are not well-defined for continuous image outputs, the decoding step from GCNN embedding to image is never specified, and the experimental tables are not reproducible from the given protocol. These are load-bearing issues rather than presentation problems, so the paper cannot be accepted in its current form.

major comments (5)
  1. [Section 4.4, Eq. (10)] The loss objective contains a sign error that inverts the stated diversity goal. Eq. (10) defines L(x) = C(x) + λD(x) + γL_laplace to be minimized, while the text says 'to promote diversity, the Kullback-Leibler (KL) Divergence loss needs to be maximized.' With a positive λ, minimizing L minimizes D, so the claimed accuracy-diversity trade-off is the opposite of what the text describes. The authors must either introduce a negative coefficient or reformulate D as a maximization/side objective, and this change affects the interpretation of all diversity-related claims, including Table 7.
  2. [Section 4.2, Definitions 4.4-4.5] The connectivity functions CCF and PCF are not well-defined for continuous-valued model outputs. PCF(M_i, M_j) = |Y*_i ∩ Y*_j|, where Y*_i is a union of predicted images or latents; for real-valued images the intersection is almost surely empty, making all PCF edge weights zero and the maximum spanning tree degenerate. CCF has a related issue: the specification sets in Definition 4.2 are sets of booleans, so |S_i ∩ S_j| only counts a small number of layer-type indicators and does not capture meaningful model similarity. The authors need to replace these with measurable distance or similarity functions (e.g., feature-space distances, kernel functions) and specify how the resulting scalar edge weights are computed.
  3. [Section 4.5, Eq. (12), and Algorithm 1] The pipeline from the GCNN to the final generated image is never defined. Section 4.1 says the GCNN embedding is 'processed by a fully connected layer to generate a final output image,' but the only output equation, y_i = softmax(W_out h_i^(L) + b_out), is a classification or regression vector, not an image, a latent code, or a denoising trajectory. Algorithm 1 returns a trained GCNN and does not describe how that GCNN is used at inference to produce the images scored in Tables 3-6. Without a specified decoding or aggregation mechanism, the reported FID, IS, and CLIP numbers for LGR-AD cannot be reproduced from the described method.
  4. [Section 5.2, Tables 3-6 and 8] The experimental evaluation is under-specified and contains internal inconsistencies. The main tables do not state the number of generated samples, prompts used, seeds, or the exact evaluation protocol. More seriously, Table 8 reports LGR-AD (All models) as FID 9.50, IS 143, CLIP 0.46, while Table 3 reports LGR-AD (Our) as FID 9.52, IS 129, CLIP 0.36 on the same configuration; the discrepancy is large and unexplained. Table 7 also compares 'Ours' on COCO/ViT-L with baselines on different datasets and backbones, so the diversity comparison is not controlled. These issues directly undermine the paper's central claim of consistent state-of-the-art performance.
  5. [Abstract and Section 4.4] The paper repeatedly claims a 'theoretical analysis' as a contribution, but no theorem, lemma, formal proof, or even a precise statement appears anywhere in Sections 4.3, 4.4, or 6. The text only provides informal intuition about the MST and the loss function. Either a rigorous theoretical statement must be added, or the claims of theoretical analysis should be removed or substantially weakened.
minor comments (7)
  1. [Abstract and Section 1] The method name is inconsistently written as both 'LGR-ED' and 'LGR-AD' in the contributions paragraph; the correct name should be used throughout.
  2. [Section 4.5, Eq. (12)] Equation (12) contains a typesetting error, '|+mathcalA|', which should be the number of nodes |V| or |A|; the corrupted notation makes the output dimensionality unclear.
  3. [Section 4.4, Eq. (10)] The regularization coefficients λ and γ are never assigned numerical values, and no sensitivity analysis is reported; without these values the ablations in Tables 8 and 9 cannot be reproduced.
  4. [Section 4.6, Algorithm 1] Line 8 refers to extracting k maximum spanning trees, but the text later says k=1; the algorithm should use a consistent notation, e.g., k-MST or MST, and state the chosen value in the pseudo-code.
  5. [Section 5.3, Figure 2] The qualitative comparison in Figure 2 is not reproducible because the prompts used to generate the displayed images are not listed, and it is unclear whether all baselines were run with identical sampling settings.
  6. [References] Several references are mismatched with their citations: reference [8] is not the MSCOCO dataset paper, reference [19] is not the original FID source, and the CLIP score citation [49] is a survey rather than the CLIP paper. The reference list should be cleaned up.
  7. [Section 5.1] The text says dataset descriptions can be found in the Appendix, but no appendix is present in the manuscript; the descriptions should be included or the sentence removed.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the graph-construction and GCNN training pipeline is internally inconsistent and incomplete, but no prediction reduces by construction to its inputs.

full rationale

I walked the paper's derivation chain and found no step in which a claimed prediction is, by the paper's own equations or by a self-citation chain, equivalent to its input. The graph representation in Section 4.2 is built from model outputs and specifications on training data (Eqs. 3-5), and the GCNN in Sections 4.4-4.5 is trained on that graph with the loss in Eq. (10) and produces y_i = softmax(W_out h_i^(L) + b_out) in Eq. (12). This is a standard supervised pipeline: the graph features are derived from training-data outputs, and the GCNN is trained with a task loss. There is no equation that defines the target y_i as the same training outputs used to build the graph, so the self-definitional and fitted-input-as-prediction patterns do not apply. The paper does not invoke any uniqueness theorem or load-bearing self-citation; the reference list contains no prior work by the present authors, and the cited works are external. The most serious problems are correctness and completeness issues, not circularity. First, the loss in Eq. (10) adds +lambda*D(x) to a loss to be minimized, while Section 4.4 states that the KL divergence loss 'needs to be maximized'; as written, minimizing L suppresses diversity, contradicting the claimed diversity-accuracy balance. Second, the inference-time decoding step is never specified: Section 4.1 says the fully connected layer 'generate[s] a final output image,' but the only output equation is a softmax over c classes, and Algorithm 1 trains and returns a GCNN without defining how images in Tables 3-6 are produced from text prompts. These are internal gaps that make the empirical claims unsupported, but they are not reductions of a prediction to its input by construction. Therefore, under the requested standard requiring a quoted equation or self-citation that forces the result, the correct finding is no significant circularity, score 0.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central method depends on graph construction choices, an assumed diversity mechanism, and an unspecified image synthesis step. The loss function's KL term has an inconsistent sign, and the connectivity functions are not well-defined for continuous outputs.

free parameters (3)
  • lambda
    Regularization weight for KL divergence term in Eq. (10); chosen by hand, no value given.
  • gamma
    Regularization weight for Laplacian loss in Eq. (10); chosen by hand, no value given.
  • k (number of MSTs) = 1
    Only one maximum spanning tree is sampled, reducing the graph to a tree; this is an ad hoc choice (Section 4.6).
assumptions (4)
  • domain assumption Maximum spanning tree selection preserves the most relevant relationships among agents
    Stated in Section 4.3 without proof; the optimality of MST for agent selection is assumed.
  • ad hoc to paper The fully connected layer on GCNN embeddings can produce a valid final image
    Section 4.5 defines y_i as a softmax output but never explains how this becomes an image.
  • ad hoc to paper The positive coefficient on KL divergence in Eq. (10) encourages diversity
    Section 4.4 says KL should be maximized, but the loss is minimized, so the sign is inconsistent.
  • domain assumption The intersection of model outputs Y*_i in PCF is meaningful
    Definition 4.5 defines PCF as the size of the intersection of output sets; for continuous image outputs this intersection is almost surely empty.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Learning Graph Representation of Agent Diffusers." pith.science (2026). https://pith.science/paper/ZF5K5AMI

@misc{pith2026250506761,
  author       = {Pith},
  title        = {Pith review of: Learning Graph Representation of Agent Diffusers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZF5K5AMI}},
  note         = {Machine review of arXiv:2505.06761}
}
abstract

Diffusion-based generative models have significantly advanced text-to-image synthesis, demonstrating impressive text comprehension and zero-shot generalization. These models refine images from random noise based on textual prompts, with initial reliance on text input shifting towards enhanced visual fidelity over time. This transition suggests that static model parameters might not optimally address the distinct phases of generation. We introduce LGR-AD (Learning Graph Representation of Agent Diffusers), a novel multi-agent system designed to improve adaptability in dynamic computer vision tasks. LGR-AD models the generation process as a distributed system of interacting agents, each representing an expert sub-model. These agents dynamically adapt to varying conditions and collaborate through a graph neural network that encodes their relationships and performance metrics. Our approach employs a coordination mechanism based on top-$k$ maximum spanning trees, optimizing the generation process. Each agent's decision-making is guided by a meta-model that minimizes a novel loss function, balancing accuracy and diversity. Theoretical analysis and extensive empirical evaluations show that LGR-AD outperforms traditional diffusion models across various benchmarks, highlighting its potential for scalable and flexible solutions in complex image generation tasks. Code is available at: https://github.com/YousIA/LGR_AD

Figures

Figures reproduced from arXiv: 2505.06761 by the authors.

Figure 1
Figure 1. LGR-AD begins by training diffusion-based models for text-to-image generation, treating each model as an agent in a multi-agent system. After training each agent diffuser , we use the agents’ outputs and specifications to construct a graph, where nodes represent agents and edges capture their interactions. GCNN is then applied to learn an optimal representation of the graph, leveraging agent collaboration. The learn… view at source ↗
Figure 2
Figure 2. Selected qualitative results of LGR-AD compared to the baseline solutions [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

60 extracted references · 32 canonical work pages

  1. [1]

    Rouqaiah Al-Refai and Karthik Nandakumar. 2023. A unified model for face matching and presentation attack detection using an ensemble of vision trans- former features. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 662–671

    work page 2023

  2. [2]

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

    arXiv 2022

  3. [3]

    Matyáš Boháček and Hany Farid. 2023. A geometric and photometric exploration of gan and diffusion synthesized faces. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 874–883

    work page 2023

  4. [4]

    Minwoo Byeon, Beomhee Park, Haecheon Kim, Sungjun Lee, Woonhyuk Baek, and Saehoon Kim. 2022. COYO-700M: Image-Text Pair Dataset. https://labelbox. com/datasets/coyo-700m-image-text-pair-dataset/

    work page 2022

  5. [5]

    Chenjie Cao, Yunuo Cai, Qiaole Dong, Yikai Wang, and Yanwei Fu. 2024. LeftRefill: Filling Right Canvas based on Left Reference through Generalized Text-to-Image Diffusion Model. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 7705–7715

    work page 2024

  6. [6]

    Hanqun Cao, Cheng Tan, Zhangyang Gao, Yilun Xu, Guangyong Chen, Pheng- Ann Heng, and Stan Z Li. 2024. A survey on generative diffusion models. IEEE Transactions on Knowledge and Data Engineering (2024)

    2024

  7. [7]

    Shoufa Chen, Peize Sun, Yibing Song, and Ping Luo. 2023. Diffusiondet: Diffusion model for object detection. InProceedings of the IEEE/CVF International Conference on Computer Vision. 19830–19843

    2023

  8. [8]

    Kyunghyun Cho, Bart Van Merriënboer, Caglar Gulcehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk, and Yoshua Bengio. 2014. Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv preprint arXiv:1406.1078 (2014)

    arXiv 2014

Show all 60 references
  1. [9]

    Min Jin Chong and David Forsyth. 2020. Effectively unbiased fid and inception score and where to find them. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition . 6070–6079

    2020

  2. [10]

    Hyungjin Chung, Jeongsol Kim, Sehui Kim, and Jong Chul Ye. 2023. Parallel diffusion models of operator and image for blind inverse problems. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 6059– 6069

    2023

  3. [11]

    Florinel-Alin Croitoru, Vlad Hondru, Radu Tudor Ionescu, and Mubarak Shah

  4. [12]

    Katherine Crowson, Stella Biderman, Daniel Kornis, Dashiell Stander, Eric Halla- han, Louis Castricato, and Edward Raff. 2022. Vqgan-clip: Open domain image generation and editing with natural language guidance. In European Conference on Computer Vision. Springer, 88–105

    2022

  5. [13]

    Rohit Gandikota, Joanna Materzynska, Jaden Fiotto-Kaufman, and David Bau

  6. [14]

    Guy Gilboa, Nir Sochen, and Yehoshua Y Zeevi. 2002. Forward-and-backward diffusion processes for adaptive image enhancement and denoising. IEEE trans- actions on image processing 11, 7 (2002), 689–703

    2002

  7. [15]

    In Proceedings of the IEEE/CVF International Conference on Computer Vision

    Erasing concepts from diffusion models. In Proceedings of the IEEE/CVF International Conference on Computer Vision . 2426–2436

  8. [16]

    Jonathan Ho, Chitwan Saharia, William Chan, David J Fleet, Mohammad Norouzi, and Tim Salimans. 2022. Cascaded diffusion models for high fidelity image generation. Journal of Machine Learning Research 23, 47 (2022), 1–33

    2022

  9. [17]

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. 2020. Denoising diffusion probabilistic models. Advances in neural information processing systems 33 (2020), 6840–6851

    2020

  10. [18]

    Yuanfeng Ji, Zhe Chen, Enze Xie, Lanqing Hong, Xihui Liu, Zhaoqiang Liu, Tong Lu, Zhenguo Li, and Ping Luo. 2023. Ddp: Diffusion model for dense visual prediction. In Proceedings of the IEEE/CVF International Conference on Computer Vision. 21741–21752

    2023

  11. [19]

    Drew A Hudson, Daniel Zoran, Mateusz Malinowski, Andrew K Lampinen, Andrew Jaegle, James L McClelland, Loic Matthey, Felix Hill, and Alexander Lerchner. 2024. Soda: Bottleneck diffusion models for representation learning. In Proceedings of the IEEE/CVF Conference on Computer V...

    2024

  12. [20]

    Subhadeep Koley, Ayan Kumar Bhunia, Aneeshan Sain, Pinaki Nath Chowd- hury, Tao Xiang, and Yi-Zhe Song. 2024. Text-to-Image Diffusion Models are Great Sketch-Photo Matchmakers. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 16826–16837

    2024

  13. [21]

    Steffen Jung and Margret Keuper. 2021. Internalized biases in fréchet inception distance. In NeurIPS 2021 Workshop on Distribution Shifts: Connecting Methods and Applications

    2021

  14. [22]

    Charles Laroche, Andrés Almansa, and Eva Coupete. 2024. Fast diffusion em: a diffusion model for blind inverse problems with application to deconvolution. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 5271–5281

    2024

  15. [23]

    Wei-Sheng Lai, Jia-Bin Huang, Narendra Ahuja, and Ming-Hsuan Yang. 2017. Deep laplacian pyramid networks for fast and accurate super-resolution. In Proceedings of the IEEE conference on computer vision and pattern recognition . 624–632

    2017

  16. [24]

    Michael S Matena and Colin A Raffel. 2022. Merging models with fisher-weighted averaging. Advances in Neural Information Processing Systems 35 (2022), 17703– 17716

    2022

  17. [25]

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

    2023

  18. [26]

    Chong Mou, Xintao Wang, Liangbin Xie, Yanze Wu, Jian Zhang, Zhongang Qi, and Ying Shan. 2024. T2i-adapter: Learning adapters to dig out more controllable ability for text-to-image diffusion models. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 38. 4296–4304

    2024

  19. [27]

    Shoya Matsumori, Yuki Abe, Kosuke Shingyouchi, Komei Sugiura, and Michita Imai. 2021. LatteGAN: Visually Guided Language Attention for Multi-Turn Text-Conditioned Image Manipulation. IEEE Access 9 (2021), 160521–160532. https://doi.org/10.1109/access.2021.3129215

    2021

  20. [28]

    Liang Peng, Yujie Mo, Jie Xu, Jialie Shen, Xiaoshuang Shi, Xiaoxiao Li, Heng Tao Shen, and Xiaofeng Zhu. 2023. GRLC: Graph representation learning with con- straints. IEEE Transactions on Neural Networks and Learning Systems (2023)

    2023

  21. [29]

    Alexander Quinn Nichol and Prafulla Dhariwal. 2021. Improved denoising diffu- sion probabilistic models. In International conference on machine learning . PMLR, 8162–8171

    2021

  22. [30]

    Aditya Ramesh, Prafulla Dhariwal, Alex Nichol, Casey Chu, and Mark Chen

  23. [31]

    Jordi Pont-Tuset, Jasper Uijlings, Soravit Changpinyo, Radu Soricut, and Vittorio Ferrari. 2020. Connecting vision and language with localized narratives. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part V 16 . Springer...

    2020

  24. [32]

    Chitwan Saharia, William Chan, Saurabh Saxena, Lala Li, Jay Whang, Emily L Denton, Kamyar Ghasemipour, Raphael Gontijo Lopes, Burcu Karagol Ayan, Tim Salimans, et al. 2022. Photorealistic text-to-image diffusion models with deep language understanding. Advances in neural infor...

    2022

  25. [33]

    Christoph Schuhmann, Romain Beaumont, Richard Vencu, Cade Gordon, Ross Wightman, Mehdi Cherti, Theo Coombes, Aarush Katta, Clayton Mullis, Mitchell Wortsman, Patrick Schramowski, Srivatsa Kundurthy, Katherine Crowson, Ludwig Schmidt, Robert Kaczmarczyk, and Jenia Jitsev. 2022....

    2022 arXiv

  26. [34]

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

    2022

  27. [35]

    Jing Shi, Wei Xiong, Zhe Lin, and Hyun Joon Jung. 2024. Instantbooth: Personal- ized text-to-image generation without test-time finetuning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 8543–8552

    2024

  28. [36]

    Haoru Tan, Chuang Wang, Sitong Wu, Xu-Yao Zhang, Fei Yin, and Cheng-Lin Liu

  29. [37]

    Christoph Schuhmann, Richard Vencu, Romain Beaumont, Robert Kaczmarczyk, Clayton Mullis, Aarush Katta, Theo Coombes, Jenia Jitsev, and Aran Komatsuzaki

  30. [38]

    Catherine Wah, Steve Branson, Peter Welinder, Pietro Perona, and Serge Belongie

  31. [39]

    Bram Wallace, Meihua Dang, Rafael Rafailov, Linqi Zhou, Aaron Lou, Senthil Purushwalkam, Stefano Ermon, Caiming Xiong, Shafiq Joty, and Nikhil Naik. 2024. Diffusion model alignment using direct preference optimization. InProceedings of the IEEE/CVF Conference on Computer Visio...

    2024

  32. [40]

    Cong Wang, Kuan Tian, Yonghang Guan, Jun Zhang, Zhiwei Jiang, Fei Shen, Xiao Han, Qing Gu, and Wei Yang. 2024. Ensembling Diffusion Models via Adaptive Feature Aggregation. arXiv preprint arXiv:2405.17082 (2024)

    2024 arXiv

  33. [41]

    Weihao Xia, Yujiu Yang, Jing-Hao Xue, and Baoyuan Wu. 2021. Tedigan: Text- guided diverse face image generation and manipulation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition . 2256–2265

    2021

  34. [42]

    Tim Van Erven and Peter Harremos. 2014. Rényi divergence and Kullback-Leibler divergence. IEEE Transactions on Information Theory 60, 7 (2014), 3797–3820

    2014

  35. [43]

    Yifei Xu, Xiaolong Xu, Honghao Gao, and Fu Xiao. 2024. SGDM: An Adaptive Style-Guided Diffusion Model for Personalized Text to Image Generation. IEEE Transactions on Multimedia (2024)

    2024

  36. [44]

    Ling Yang, Jingwei Liu, Shenda Hong, Zhilong Zhang, Zhilin Huang, Zheming Cai, Wentao Zhang, and Bin Cui. 2024. Improving diffusion-based image synthesis with context prediction. Advances in Neural Information Processing Systems 36 (2024)

    2024

  37. [45]

    Ling Yang, Zhilong Zhang, Yang Song, Shenda Hong, Runsheng Xu, Yue Zhao, Wentao Zhang, Bin Cui, and Ming-Hsuan Yang. 2023. Diffusion models: A comprehensive survey of methods and applications. Comput. Surveys 56, 4 (2023), 1–39

    2023

  38. [46]

    Haonan Yin, Guanlong Jiao, Qianhui Wu, Borje F Karlsson, Biqing Huang, and Chin Yew Lin. 2023. Lafite: Latent diffusion model with feature editing for unsupervised multi-class anomaly detection. arXiv preprint arXiv:2307.08059 (2023)

    2023 arXiv

  39. [47]

    Sihyun Yu, Kihyuk Sohn, Subin Kim, and Jinwoo Shin. 2023. Video probabilis- tic diffusion models in projected latent space. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 18456–18466

    2023

  40. [48]

    Tao Xu, Pengchuan Zhang, Qiuyuan Huang, Han Zhang, Zhe Gan, Xiaolei Huang, and Xiaodong He. 2018. Attngan: Fine-grained text to image generation with attentional generative adversarial networks. In Proceedings of the IEEE conference on computer vision and pattern recognition ....

    2018

  41. [49]

    Chenshuang Zhang, Chaoning Zhang, Mengchun Zhang, and In So Kweon

  42. [50]

    Han Zhang, Jing Yu Koh, Jason Baldridge, Honglak Lee, and Yinfei Yang. 2021. Cross-modal contrastive learning for text-to-image generation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition . 833–842

    2021

  43. [51]

    Jing Zhao, Heliang Zheng, Chaoyue Wang, Long Lan, and Wenjing Yang. 2023. MagicFusion: Boosting Text-to-Image Generation Performance by Fusing Diffu- sion Models. In Proceedings of the IEEE/CVF International Conference on Computer Vision. 22592–22602

    2023

  44. [52]

    Da-Wei Zhou, Hai-Long Sun, Han-Jia Ye, and De-Chuan Zhan. 2024. Expandable subspace ensemble for pre-trained model-based class-incremental learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 23554–23564

    2024

  45. [53]

    Minfeng Zhu, Pingbo Pan, Wei Chen, and Yi Yang. 2019. Dm-gan: Dynamic mem- ory generative adversarial networks for text-to-image synthesis. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition . 5802–5810

    2019

  46. [54]

    Yu Zeng, Vishal M Patel, Haochen Wang, Xun Huang, Ting-Chun Wang, Ming-Yu Liu, and Yogesh Balaji. 2024. JeDi: Joint-Image Diffusion Models for Finetuning- Free Personalized Text-to-Image Generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recog...

    2024

  47. [56]

    arXiv preprint arXiv:2303.07909 (2023)

    Text-to-image diffusion models in generative ai: A survey. arXiv preprint arXiv:2303.07909 (2023)

    2023 arXiv

  48. [2011]

    The caltech-ucsd birds-200-2011 dataset. (2011)

    2011

  49. [2021]

    arXiv:2111.02114 [cs.CV] https://arxiv.org/abs/2111.02114

    LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs. arXiv:2111.02114 [cs.CV] https://arxiv.org/abs/2111.02114

    arXiv

  50. [2022]

    arXiv preprint arXiv:2204.06125 1, 2 (2022), 3

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

    2022 arXiv

  51. [2023]

    IEEE Transactions on Pattern Analysis and Machine Intelligence (2023)

    Diffusion models in vision: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence (2023)

    2023

  52. [2024]

    International Journal of Computer Vision (2024), 1–23

    Ensemble quadratic assignment network for graph matching. International Journal of Computer Vision (2024), 1–23

    2024

Pith tools

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