The DOI matches its Crossref record; 61 of the 61 references it cites were found in a registry or an index; 1 of the 5 sources its claims cite were checked for support.
Scalable Diffusion Models with Transformers
Open accessAbstract
We explore a new class of diffusion models based on the transformer architecture. We train latent diffusion models of images, replacing the commonly-used U-Net backbone with a transformer that operates on latent patches. We analyze the scalability of our Diffusion Transformers (DiTs) through the lens of forward pass complexity as measured by Gflops. We find that DiTs with higher Gflops—through increased transformer depth/width or increased number of input tokens—consistently have lower FID. In addition to possessing good scalability properties, our largest DiT-XL/2 models outperform all prior diffusion models on the class-conditional ImageNet 512×512 and 256×256 benchmarks, achieving a state-of-the-art FID of 2.27 on the latter.
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
Diffusion Transformers (DiTs) replace the standard U-Net backbone in latent diffusion models with a transformer operating on latent patches. DiT-XL/2 achieves state-of-the-art class-conditional ImageNet generation, reaching FID 2.27 at 256×256 and 3.04 at 512×512.
Key findings
- DiT block conditioning matters: adaLN-Zero outperforms cross-attention and in-context conditioning at all training stages while being the most compute-efficient; at 400K iterations its FID is nearly half that of the in-context model.
- Scaling DiT Gflops improves sample quality: across 12 models, transformer Gflops and FID-50K at 400K steps have a correlation of −0.93, with gains from increasing depth/width or number of input tokens.
- DiT-XL/2 sets state of the art: FID 2.27 on class-conditional ImageNet 256×256 with classifier-free guidance cfg=1.50, improving on LDM’s previous best 3.60; at 512×512 it reaches FID 3.04 versus ADM’s 3.85.
- Larger DiT models are more compute-efficient, and extra sampling compute cannot compensate for smaller models: DiT-L/2 with 1000 sampling steps uses 80.7 Tflops and gets FID-10K 25.9, while DiT-XL/2 with 128 steps uses 15.2 Tflops and gets FID-10K 23.7.
Methods
- Design: latent diffusion with a frozen off-the-shelf Stable Diffusion VAE (downsample factor 8); DiT patchifies latents, applies ViT-style positional embeddings, processes tokens through N transformer blocks, and linearly decodes to noise and covariance predictions. Design space includes patch sizes p=2,4,8; block variants in-context, cross-attention, adaLN, and adaLN-Zero; model sizes DiT-S/B/L/XL.
- Data and training: class-conditional ImageNet at 256×256 and 512×512; AdamW, constant learning rate 1×10⁻⁴, batch size 256, horizontal flips, EMA decay 0.9999, no learning-rate warmup or regularization; diffusion hyperparameters retained from ADM; implemented in JAX on TPU-v3 pods.
- Evaluation: FID-50K with 250 DDPM sampling steps using ADM’s TensorFlow evaluation suite; 12 DiT models swept over S/B/L/XL and p=8/4/2 at 400K steps; DiT-XL/2 trained for 7M steps at 256×256 and 3M steps at 512×512; additional comparisons of conditioning strategies and model versus sampling compute.
Limitations
- The paper does not state an explicit limitations section.
- Evident from the text: experiments are limited to class-conditional ImageNet at 256×256 and 512×512; text-to-image and other domains are left to future work.
- Evident from the text: DiTs use a frozen off-the-shelf VAE, so generation quality depends on that VAE; training hyperparameters were retained from ADM and not tuned; pixel-space DiT is mentioned as possible but not tested.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
28 Own result · 5 Own interpretation · 4 No basis given · 3 Sources not found · 2 Own method · 1 Partly supported · 1 Not checked yet · accuracy not yet measured
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“As shown in Figure 2, LDMs achieve good performance while using a fraction of the Gflops of pixel space diffusion models like ADM.”
§ 3. Diffusion Transformers > 3.1. Preliminaries
What the sources say
- Partly supportedcites [48] sourceExperimental — accuracy not yet measured
Section 4.2 reports performance close to ADM on LSUN-Bedrooms while using four times fewer training resources, and Appendix D.4 describes competitive ImageNet performance with lower compute requirements. These passages support the performance-and-efficiency comparison, but our copy does not give a GFLOPs comparison with ADM; it reports training resources instead. (LLM, conf=0.85)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Larger DiT models are more compute-efficient.”
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Adaptive layer norm works best.”
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Transformer Gflops are strongly correlated with FID.”
boostedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Performance steadily improves in FID as model flops increase.”
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Increasing transformer forward pass Gflops increases sample quality.”
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The results demonstrate that different DiT configs obtain similar FID values when their total Gflops are similar (e.g., DiT-S/2 and DiT-B/4).”
boostedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Indeed, scaling both model size and the number of tokens yields notable improvements in visual quality.”
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Across all four configs, significant improvements in FID are obtained over all stages of training by making the transformer deeper and wider.”
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Scaling the transformer backbone yields better generative models across all model sizes and patch sizes.”
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- The quote only partly states this claim. 4
- Probably not a substantial claim. 6
- It may state two separate claims as one. 18
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Our best model, DiT-XL/2, is compute-efficient and outperforms all prior U-Net-based diffusion models, like ADM and LDM.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The quote only partly states this claim.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Cross-attention adds the most Gflops to the model, roughly a 15% overhead.”
§ 3. Diffusion Transformers > 3.2. Diffusion Transformer Design Space
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“As with prior work using guidance, we observe that larger scales increase visual fidelity and decrease sample diversity.”
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For both the 256 × 256 and 512 × 512 DiT-XL/2 models, we never observed FID saturate and continued training them as long as possible.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Of the three block designs we explore, adaLN adds the least Gflops and is thus the most compute-efficient.”
§ 3. Diffusion Transformers > 3.2. Diffusion Transformer Design Space
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Our best model, DiT-XL/2, is compute-efficient and outperforms all prior U-Net-based diffusion models, like ADM and LDM.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Table 2. Benchmarking class-conditional image generation on ImageNet 256×256. DiT-XL/2 achieves state-of-the-art FID.”
§ 107 108 109 1010 1011 1012 > 5.1. State-of-the-Art Diffusion Models
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one. The quote only partly states this claim.
- Rests on an assumption: FID-50K values reported by prior work are directly comparable to the FID-50K values computed in this paper.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“When trained for only 2.35M steps (similar to ADM), XL/2 still outperforms all prior diffusion models with an FID of 2.55.”
§ 107 108 109 1010 1011 1012 > 5.1. State-of-the-Art Diffusion Models
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In all cases, we find that increasing model size and decreasing patch size yields considerably improved diffusion models.”
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“XL/2 again outperforms all prior diffusion models at this resolution, improving the previous best FID of 3.85 achieved by ADM to 3.04.”
§ 107 108 109 1010 1011 1012 > 5.1. State-of-the-Art Diffusion Models
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Rests on an assumption: FID-50K values reported by prior work are directly comparable to the FID-50K values computed in this paper.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“we did not find learning rate warmup nor regularization necessary to train DiTs to high performance.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Increasing the Gflops in the model—either by increasing transformer depth/width or increasing the number of input tokens—yields significant improvements in visual fidelity.”
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In all cases, we find that increasing model size and decreasing patch size yields considerably improved diffusion models.”
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The quote only partly states this claim.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We find a strong negative correlation between model Gflops and FID-50K, suggesting that additional model compute is the critical ingredient for improved DiT models.”
- Also stated in: § 100 101 102 · boosted
“Transformer Gflops are strongly correlated with FID.”
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Also stated in: § 100 101 102 · boosted
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Finally, we also observe that DiT-XL/2 achieves higher recall values at all tested classifier-free guidance scales compared to LDM-4 and LDM-8.”
§ 107 108 109 1010 1011 1012 > 5.1. State-of-the-Art Diffusion Models
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The quote only partly states this claim.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Increasing DiT model Gflops (via transformer size or number of input tokens) causes the training loss to decrease more rapidly and saturate at a lower value.”
- Also stated in: § 0 10K 20K 30K 40K 50K 60K 70K 80K 90K 100K · plain
“Note that scaled-up DiT models exhibit lower training losses.”
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Also stated in: § 0 10K 20K 30K 40K 50K 60K 70K 80K 90K 100K · plain
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“We did not tune learning rates, decay/warm-up schedules, Adam β1/β2 or weight decays.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“which have been shown to scale more effectively for visual recognition than traditional convolutional networks (e.g., ResNet [15]).”
What the sources say
- Not matchedcites [10] Dosovitskiy, An image is worth 16x16 words: Transformers for image recognition at scale
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
- Not matched
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“Even without these techniques, training was highly stable across all model configs and we did not observe any loss spikes commonly seen when training transformers.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“diffusion models have been at the forefront of recent advances in image-level generative models [9,46]; yet, they all adopt a convolutional U-Net architecture as the de-facto choice of backbone.”
What the sources say
- Not matchedcites [9] Dhariwal, Diffusion models beat gans on image synthesis
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites [46] Ramesh, 2022, Hierarchical text-conditional image generation with clip latents
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
In the paper
“Note that prior work [9] measures Precision and Recall using 1000 real samples for 512 × 512 resolution; for consistency, we do the same.”
§ 107 108 109 1010 1011 1012 > 5.1. State-of-the-Art Diffusion Models
What the sources say
- Not matchedcites [9] Dhariwal, Diffusion models beat gans on image synthesis
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plainn = 1,000- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“halving pwill quadruple T , and thus at least quadruple total transformer Gflops.”
§ 3. Diffusion Transformers > 3.2. Diffusion Transformer Design Space
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“A smaller patch size p results in a longer sequence length and thus more Gflops.”
§ 3. Diffusion Transformers > 3.2. Diffusion Transformer Design Space
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
The paper relies on this without support.
FID-50K values reported by prior work are directly comparable to the FID-50K values computed in this paper.
“We show samples from the model in Figures 1, and we compare against state-of-the-art class-conditional generative models. We report results in Table 2.”
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“Unlike prior works that employ text-to-image (T2I) U-Net [10, 33, 35, 40] for talking head generation, MoCha is built on a diffusion transformer (DiT) [26]. By incorporating text and speech conditions sequentially via cross-attention, it effectively captures both semantics and temporal dynamics.”
From MoCha: Towards Movie-Grade Talking Character Synthesis (2025) · cites this paper as [26]
In the citing paper
“Supervised learning, a foundational approach in AI, uses labeled datasets to guide models towards accurate predictions, and it has been integral to various applications, including image recognition and NLP [163], [164], [165].”
From From Google Gemini to OpenAI Q* (Q-Star): A Survey on Reshaping the Generative Artificial Intelligence (AI) Research Landscape (2025) · cites this paper as [164]
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
U-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Image-to-Image Translation with Conditional Adversarial Networks
Isola, P., Zhu, J.-Y., Zhou, T., & Efros, A. A. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.632 · Cited by 22,930
High-Resolution Image Synthesis with Latent Diffusion Models
Rombach, R., Blattmann, A., Lorenz, D., Esser, P., & Ommer, B. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01042 · Cited by 20,302
Auto-Encoding Variational Bayes
Kingma, D. P., & Welling, M. (2013) · arXiv · doi:10.48550/arxiv.1312.6114 · Cited by 20,067
Research works printed but not linked: 3
- Chen, Generative pre-training from pixels
- Dosovitskiy, An image is worth 16x16 words: Transformers for image recognition at scale
- Goodfellow, Generative adversarial nets
Cited by
Sorted by citations
Items 2,301–2,400 of 3,684
Amber-Image: Efficient Compression of Large-Scale Diffusion TransformersPreprint
Yang, C., Li, T., Zhang, Y., & Gao, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2602.17047 · Cited by 1
Su, C., & Emad, A. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.02.25.707938 · Cited by 1
LIPAR: Training-free Latent Inter-Frame Pruning with Attention RecoveryPreprint
Menn, D., Yang, Y., Wang, B., Wei, X., Munir, M., Liang, F., Mărculescu, R., Xu, C., & Marculescu, D. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.05811 · Cited by 1
Too Vivid to Be Real? Benchmarking and Calibrating Generative Color FidelityPreprint
Fang, Z., Jia, Z., Zhong, Y., Luo, P., Zhang, J., Lu, G., Yu, J., & Pei, W. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.10990 · Cited by 1
Warm-Start Flow Matching for Guaranteed Fast Text/Image GenerationPreprint
Kim, M. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.19360 · Cited by 1
OmniDiT: Extending Diffusion Transformer to Omni-VTON FrameworkPreprint
Zeng, W., Wei, P., Wang, H., Zhang, B., Sun, J., Fan, D., Lin, H. E., Chen, L., Gan, Q., Yang, F., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.19643 · Cited by 1
SARe: Structure-Aware Generative 3D Fragment ReassemblyPreprint
Jia, H., Wang, C., Yang, Y., Jiang, Z., Luo, Y., Ye, S., & Tang, T. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.21611 · Cited by 1
Three Creates All: You Only Sample 3 StepsPreprint
Cai, Y., Wang, G., Li, Z., Li, L. E., Liu, Z., & Su, S. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.22375 · Cited by 2
Not All Denoising Steps Are Equal: Model Scheduling for Faster Masked Diffusion Language ModelsPreprint
Sedykh, I., Sorokin, N., & Malykh, V. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.02340 · Cited by 1
LiveStre4m: Feed-Forward Live Streaming of Novel Views from Unposed Multi-View VideoPreprint
Quesado, P., Akdag, E., Kashefbahrami, Y., Menu, W. J., & Bondarev, E. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.06740 · Cited by 1
Attention GeometryPreprint
Candanedo, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.09560 · Cited by 1
Oracle Noise: Faster Semantic Spherical Alignment for Interpretable Latent OptimizationPreprint
Li, H., Chen, W., Wang, L., Liang, S., Jia, H., & Yue, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.23540 · Cited by 1
The two clocks and the innovation window: When and how generative models learn rulesPreprint
Wang, B., Finn, E. L. B., & Liu, B. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.10019 · Cited by 1
MoZoo:Unleashing Video Diffusion power in animal fur and muscle simulationPreprint
Liu, D., Ma, J., Yang, X., Zhang, J., Xia, B., Kan, Z., Huang, N., Liang, J., Yang, W., & Li, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.13857 · Cited by 1
DSSP: Diffusion State Space Policy with Full-History EncodingPreprint
Guan, Z., Hu, J., Fang, H., Jiang, Y., Huang, Y., Li, S., Li, X., & Ban, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.14598 · Cited by 1
Composition of Memory Experts for Diffusion World ModelsPreprint
Stapf, S., Acuaviva, P., Davtyan, A., & Favaro, P. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.18813 · Cited by 1
FlowLong: Inference-time Long Video Generation via Manifold-constrained Tweedie MatchingPreprint
Park, J., Park, G. Y., Kwon, G., & Ye, J. C. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.20910 · Cited by 1
Veda: Scalable Video Diffusion via Distilled Sparse AttentionPreprint
Han, S., Yang, H., Hu, X., Mei, X., Jiang, Y., & Qi, X. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.30325 · Cited by 1
Co-training with Ego-centric Video and Demonstration for Robot Navigation TaskPreprint
Kuno, S., Ouchi, Y., & Suzuki, K. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.01951 · Cited by 1
GLENS: Global Search via Learning from Solver Iterates with Diffusion ModelsPreprint
Li, A., Stellato, B., & Beeson, R. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.00366 · Cited by 1
Jiang, C., Qian, S., Pedro, K., Amram, O., Qu, H., & Voetberg, M. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.04165 · Cited by 1
Activation Steering of Video Generation Models via Reduced-Order Linear Optimal ControlPreprint
Hong, J., Chan, A., Dai, Q., Skifstad, J., & Chou, G. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.04775 · Cited by 1
Gandhi, D. B., Asaria, A., & Salomone, T. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.12280 · Cited by 1
GF-DiT: Scheduling Parallelism for Diffusion Transformer ServingPreprint
Qiang, X., Hu, Y., Sun, S., Yang, J., Zhao, H., Chen, C., Feng, Y., Leng, J., & Guo, M. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.13501 · Cited by 1
The Geometry of Phase Transitions in Generative Dynamics via Projection CausticsPreprint
Sakamoto, R., & Sakamoto, K. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.13191 · Cited by 1
Unified Motion-Action Modeling for Heterogeneous Robot LearningPreprint
Cao, Y., Liu, S., Feng, C., Zhang, M., Lu, X., Owens, A., & Fang, K. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.16917 · Cited by 1
Zhang, B., Wang, Y., Hu, D., Fang, J., Liu, Y., Zhao, C., Li, D., Lan, D., Liu, Y., Li, X., et al. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6962791 · Cited by 1
Remember what you did?: Learning Behavioral Memories for Partially Observable Object ManipulationPreprint
Wang, K., Yeom, S., Cao, J., Zhi, Y., Shinde, N., & Yip, M. C. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.21188 · Cited by 1
Unbalanced Perturbation Dynamics For Cell Fate DesignPreprint
Peng, Q., Wang, Y., Li, J., Wang, X., Xiao, Y., & Zhou, P. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.06.30.735555 · Cited by 1
Training Hybrid Block Diffusion Language Models with Partial BidirectionalityPreprint
Chaturvedi, P., Shroff, P., Suresh, T., Kang, H., & Wen, K. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.02805 · Cited by 1
Xema: Efficient Diffusion Serving through Fine-Grained Memory Management and Auto-ConfigurationPreprint
Kang, X., Xiang, G., Li, S., Wang, Y., Shi, S., Zhang, L., & Chu, X. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.11136 · Cited by 2
FlashDiff: Efficient Regional Execution and Scheduling for Diffusion Model ServingPreprint
Qiao, Y., He, P., Xie, S., Barik, A., Zhang, C., Tu, Z., & Lai, F. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.12121 · Cited by 1
Self-citationZhang, Y., Wu, Y., Xiao, Y. H., Xue, C., Ren, Y., Li, C., Yin, Y., Niu, D., & Sun, G. X. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.14908 · Cited by 1
Learning Standard Model structure from LHC data with Riemannian flow matchingPreprint
Kato, M., Urquía-Calderón, K. A., Timiryasov, I., & Oleg Ruchayskiy, C. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.16144 · Cited by 1
Image Editing Models are Numerical SolversPreprint
Mizrahi, U. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.18787 · Cited by 1
Learning Sampling Parameters for Diffusion ModelsPreprint
Lim, A., & Gandelsman, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.23488 · Cited by 1
Latent-Kernel Discrete Flow Maps for Few-Step GenerationPreprint
Ahmed, M., Fan, Y.-T., Venkateswara, H., & Patterson, M. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.27529 · Cited by 1
Using Diffusion Models to Estimate Uncertainties in Analytic ContinuationPreprint
Meir, S., Freedman, D., & Hirshberg, B. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.13123 · Cited by 1
Equilibrium Forcing: Adaptive Video Generation Without Noise ConditioningPreprint
Lillemark, H. J., Rojas, A., Novack, Z., Wang, R., Du, Y., Ma, Y., Berg-Kirkpatrick, T., & Yu, R. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.14706 · Cited by 1
JoLT: Joint Latent Trajectories for Context-Guided High-Resolution Tiled GenerationPreprint
Koroglu, M., Jeanneret, G., Caselles-Dupré, H., Cord, M., & Dapogny, A. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.15395 · Cited by 1
Hierarchical Prototype Emergence in Modern Hopfield ModelsPreprint
Cowsik, A., & Sriram, A. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.12079 · Cited by 1
ZYT-World: A Real-Time Controllable World Model for Closed-Loop Autonomous-Driving SimulationPreprint
Hu, B., Wei, X., Huang, H., Huang, Y., Wang, C., Yang, Y., Wang, J., Zhu, R., Yang, Z., Liu, G., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.21712 · Cited by 1
Applications of Neural Cellular Automata: State of the Art, Challenges and OpportunitiesPreprint
Lemke, N., Ihm, N., Kalkhof, J., Konstantin, M., Krumb, H., Lang, D. M., Pajouheshagar, E., Sadafi, A., Kuijper, A., Lekadir, K., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.24595 · Cited by 1
Generating Chest X-Ray Counterfactuals by Specialising Foundation Image ModelsPreprint
Xing, X., Rasal, R., Meister, J. A., Ghorayeb, S., Khara, G. S., & Schrouff, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.24879 · Cited by 1
OMatG-flash: An All-Atom Flow Map with Reinforce Adjoint Matching for Scalable Materials DiscoveryPreprint
Egg, T., Sullivan, H. W., Tadmor, E. B., & Martiniani, S. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.26402 · Cited by 1
Li, Y., Yu, J., Chen, J., Li, H., Xue, S., Liu, H., Luo, P., Han, S., & Xie, E. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.35110 · Cited by 3
Test-Time Training as Residual Memory for Robot PoliciesPreprint
Wang, H., Zhang, G., Kompella, R. R., Liu, G., & Yan, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.04701 · Cited by 1
LiFT: Loop Flow TransformersPreprint
Derakhshani, M. M., Curvo, P. M. P., Burghouts, G. J., van de Meent, J.-W., & Snoek, C. G. M. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.05538 · Cited by 1
S2PD: Serial-to-Parallel Diffusion for Physically and Logically Consistent Video GenerationPreprint
Hu, J., Reino, D. O., & Tewari, A. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.06847 · Cited by 1
Workhorse: Learning Robust Whole-Body Humanoid Loco-Manipulation from Human DataPreprint
Hu, S., Liao, Q., Chi, Y., Zakka, K., Shao, Y. S., Abbeel, P., & Sreenath, K. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.09117 · Cited by 1
Kong, Z., Li, Y., Zeng, F., Xin, L., Messica, S., Lin, X., Zhao, P., Kellis, M., Tang, H., & Zitnik, M. (2025) · arXiv · doi:10.48550/arxiv.2505.18227 · Cited by 51
“In modern generative models [11, 42, 88, 104, 107], a token denotes one fundamental unit of input or representation, typically encoded as a vector.”
MoCha: End-to-End Video Character Replacement without Structural Guidance
Xu, Z., Ma, J., Wang, Z., Peng, Z., Liang, J., & Li, J. (2026) · arXiv
“Based on the clean latent z 0 , a noisy latent z t is constructed by a randomly sampled timestep t and a noise ω sampled from standard Gaussian Distribution N (0 , 1) : During training, z t is fed to a Transformer-based diffusion model [22] (DiT).”
MoCha: Towards Movie-Grade Talking Character Synthesis
Wei, C., Sun, B., Ma, H., Hou, J., Juefei-Xu, F., He, Z., Dai, X., Zhang, L., Li, K., Hou, T., et al. (2025) · arXiv · doi:10.48550/arxiv.2503.23307
“Unlike prior works that employ text-to-image (T2I) U-Net [10, 33, 35, 40] for talking head generation, MoCha is built on a diffusion transformer (DiT) [26].”
Kwon, P., & Chen, C. (2025) · arXiv · doi:10.48550/arxiv.2512.01686 · Cited by 1
“Although recent image customization methods for diffusion transformer (DiT) models [2, 14, 38] can control who appears in the image [35, 61, 63] or where objects are placed [6, 72], they struggle to support both simultaneously, resulting in images with overlapping characters or incorrect…”
Ai, Q., Zhan, J., & Liu, Y. (2025) · arXiv · doi:10.48550/arxiv.2501.02842
“Language models are increasingly applied in the multimodal generation domain, such as in image [71, 72] and video generation [73, 74].”
SVG: 3D Stereoscopic Video Generation via Denoising Frame Matrix
Dai, P., Tan, F., Xu, Q., Futschik, D., Du, R., Fanello, S., Qi, X., & Zhang, Y. (2024) · arXiv · doi:10.48550/arxiv.2407.00367
Manga Generation via Layout-controllable Diffusion
Chen, S., Li, D., Bao, Z., Zhou, Y., Tan, L., Zhong, Y., & Zhao, Z. (2024) · arXiv · doi:10.48550/arxiv.2412.19303
LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels
Maes, L., Lidec, Q. L., Scieur, D., LeCun, Y., & Balestriero, R. (2026) · arXiv · doi:10.48550/arxiv.2603.19312
AniGen: Unified S 3 Fields for Animatable 3D Asset Generation
Huang, Y.-H., Zou, Z.-X., He, Y., Chang, C., Pu, C.-F., Yang, Z., Guo, Y.-C., Cao, Y.-P., & Qi, X. (2026) · ACM Transactions on Graphics · doi:10.1145/3811297
EMO2: End-Effector Guided Audio-Driven Avatar Video Generation
Tian, L., Hu, S., Wang, Q., Zhang, B., & Bo, L. (2025) · arXiv · doi:10.48550/arxiv.2501.10687
T1: Scaling Diffusion Probabilistic Fields to High-Resolution on Unified Visual ModalitiesPreprint
Mei, K., Zhou, M., & Patel, V. M. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2305.14674
Signal as Token: Robust DOA Estimation in Complex Environments Aidded by TransformerPreprint
Wang, Z., & Cao, Z. (2023) · doi:10.36227/techrxiv.24166209
Signal as Token: Robust DOA Estimation in Complex Environments Aidded by TransformerPreprint
Wang, Z., & Cao, Z. (2023) · doi:10.36227/techrxiv.24166209.v1
Signal as Token: Robust DOA Estimation in Complex Environments Aidded by TransformerPreprint
Wang, Z., & Cao, Z. (2023) · doi:10.36227/techrxiv.24166209.v2
Signal as Token: Robust Doa Estimation in Complex Environments Aided by TransformerPreprint
Wang, Z., Cao, Z., Xie, J., Li, H., & He, Z. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4604922
Collaborative Neural PaintingPreprint
Dall’Asen, N., Menapace, W., Peruzzo, E., Sangineto, E., Wang, Y., & Ricci, E. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2312.01800
PassDiff: A New Approach for Password Guessing Using Diffusion Model
Guo, S., Duan, M., Du, Y., Wang, W., & Guo, L. (2024) · Lecture notes in electrical engineering · doi:10.1007/978-981-99-9239-3_3
Distortion is All You Need: From Image Denoising to Quality AssessmentPreprint
Gui, Z., Liu, J., Yuan, C., Yang, G., & He, W. P. (2024) · Research Square · doi:10.21203/rs.3.rs-3893378/v1
TDS: Two Diffusion Streams for Low-Light Image Enhancement
Wang, J., Liu, X., Zhang, Y., & Hu, J. (2023) · International Conference on Software Engineering and Computer Science (CSECS) · doi:10.1109/csecs60003.2023.10428430
Lu, J., & Pan, B. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4822798
Customize Your Own Paired Data via Few-shot WayPreprint
Chen, J., Li, B., Hua, M., Xu, P., & He, Q. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2405.12490
Virtual avatar generation models as world navigatorsPreprint
Mandava, S. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2406.01056
KDD workshop on Evaluation and Trustworthiness of Generative AI Models
Ling, Y., Dong, S., Feng, Y., Liu, Z. J., Karypis, G., & Reddy, C. K. A. (2024) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3637528.3671481
CELL-Diff: Unified Diffusion Modeling for Protein Sequences and Microscopy ImagesPreprint
Dihan, Z., & Huang, B. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.10.15.618585
Multivariate activation functions
Ivaniuk, A. (2024) · International Journal of Wavelets Multiresolution and Information Processing · doi:10.1142/s0219691324500565
Chandran, P., Serifi, A., Groß, M. H., & Bächer, M. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73016-0_1
Generative Factor Chaining: Coordinated Manipulation with Diffusion-based Factor GraphPreprint
Mishra, U. A., Chen, Y., & Xu, D. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2409.16275
Implicit Dynamical Flow Fusion (IDFF) for Generative ModelingPreprint
Rezaei, M. R., Popović, M. R., Lankarany, M., & Krishnan, R. G. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2409.14599
Wei, Y., Huang, L., Wu, Z.-F., Wang, W., Yu, L., Jia, M., & Ma, S. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73220-1_2
Latent RAGE: Randomness Assessment Using Generative Entropy Models
Pratihar, K., Chakraborty, R. S., & Mukhopadhyay, D. (2024) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2024.3449562
Asteroid Image Inpainting Method Based on Accelerated Diffusion Model
Zheng, Y., & Huang, X. (2024) · International Conference on Intelligent Computing and Signal Processing (ICSP) · doi:10.1109/icsp62122.2024.10743626
Accelerated AI Inference via Dynamic Execution MethodsPreprint
Barad, H., Achterberg, J., Chou, T.-L., & Yu, J. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2411.00853
A Powder Diffraction-AI Solution for Crystalline StructurePreprint
Wu, D., Wang, P. F., Zhou, S., Zhang, B., Yu, L., Chen, X., Wang, X., Zhou, Z., Wang, Y., Wang, S., et al. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2411.06062
Hyperspectral data augmentation with transformer-based diffusion models
Ferrari, M., & Bruzzone, L. (2024) · Image and signal processing for remote sensing. Conference · doi:10.1117/12.3032957
Mask-ControlNet: Higher-Quality Image Generation with an Additional Mask Prompt
Huang, Z., Xiong, H., Wang, H., Wang, L., & Li, Z. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-78172-8_6
Accelerating Vision Diffusion Transformers with Skip BranchesPreprint
Chen, G., & Zhao, X. (2024) · Qeios · doi:10.32388/bim5v9
PSG-Adapter: Controllable Planning Scene Graph for Improving Text-to-Image Diffusion
Gao, Y. (2024) · Lecture notes in computer science · doi:10.1007/978-981-96-0917-8_12
Adversarial Purification by Consistency-Aware Latent Space Optimization on Data Manifolds
Zhang, S., Yang, J., Luo, H., Chen, J., Wang, K., Liu, F., Han, B., & Tan, M. (2026) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2026.3720823
A Fuzz Testing Method Based on DDPM for Intelligent Connected Vehicles CAN Communication
Shen, L., Xiu, J., Zhang, Z., & Yang, Z. (2024) · SAE technical papers on CD-ROM/SAE technical paper series · doi:10.4271/2024-01-7044
Conceptual Model and NLP-System "Text to Image"
Maslianko, P., & Pavlovska, K. (2024) · Cybernetics and Computer Technologies · doi:10.34229/2707-451x.24.4.9
Focal Diffusion Process for Object-Aware 3D LiDAR Generation
Zhang, H., & Liu, X. (2024) · ACM International Conference on Multimedia in Asia (MMAsia) · doi:10.1145/3696409.3700275
Synthetic Soiling Augmentation Using the Diffusion Model for Automotive Camera System
Heo, J., Kim, T., & Hwang, Y. (2024) · International Conference on Platform Technology and Service · doi:10.1109/platcon63925.2024.10830729
Faster and More Efficient Subject Image Generation for Text-to-Image Diffusion Models
Yu, B., Yao, Z., Wei, J., Sun, L., Zhang, S., & Bu, L. (2024) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc54092.2024.10831577
Zhong, C. (2025) · Applied and Computational Engineering · doi:10.54254/2755-2721/2025.20611
LoRaFlow: High-Quality Signal Reconstruction using Rectified FlowPreprint
Osman, M., & Nadeem, T. (2025) · Qeios · doi:10.32388/ku8ijy
Towards Fast Face Image Quality Assessment via Latent Diffusion Model
Yan, Z., Zhao, W., Zhu, X., Gao, L., Zhang, X., & Lei, Z. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-1068-6_25
Secure & Personalized Music-to-Video Generation via CHARCHAPreprint
Agarwal, M., Agarwal, G., Benoit, S., Lippman, A., & Oh, J. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2502.02610
Advancements and future pathway in human body digital twins
Chen, X., Tang, C., Dai, Y., Li, C., & Gao, S. (2025) · Elsevier eBooks · doi:10.1016/b978-0-443-34226-4.00019-8
Transformer-based Diffusion Model for Single Image Deblurring
Ye, Z., Yang, M., & Yang, L. (2024) · China Automation Congress (CAC) · doi:10.1109/cac63892.2024.10864714
Restoring Missing Slices of Serial Section Electron Microscopy Using Diffusion Models
Yang, H., Zhang, R., & Zhang, Q. (2025) · Communications in computer and information science · doi:10.1007/978-981-96-1904-7_1
Metadata
- Authors
- Peebles, W., & Xie, S.
- Affiliations
- University of California, Berkeley (US)
- New York University (US)
- ORCID
- Xie, Saining
- Venue
- 2023 IEEE/CVF International Conference on Computer Vision (ICCV)
- Publisher
- IEEE
- Published
- 2023-10-01
- Pages
- 4172-4182
- OpenAlex
- W4390872297
- Semantic Scholar
- 736973165f98105fec3729b7db414ae4d80fcbeb
- Cited by
- 2,284
- References
- 59
- Citations by year
- 2026: 1007 · 2025: 925 · 2024: 315 · 2023: 37
- FWCI
- 244.26
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE/CVF International Conference on Computer Vision (ICCV) (publishedVersion)
- arXiv (Cornell University) (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Generative Adversarial Networks and Image Synthesis
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image and Signal Denoising Methods
- Keywords
- diffusion models, latent diffusion models, diffusion transformer, scalability, image generation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-11-08, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-11-08 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
