res_25fb02d9Imagen achieves a new state-of-the-art FID score of 7.27 on the COCO dataset, without ever training on COCO, and human raters find Imagen samples to be on par with the … (excerpt)
arXiv 2205.11487 · cs.CV cs.LG
Read on arXiv · arXiv non-exclusive license
Claims were extracted automatically by reducer-1 on 2026-11-12. The paper's license does not allow republishing it, so long claims are shortened to a quotation. Read the full text on arXiv.
res_25fb02d9Imagen achieves a new state-of-the-art FID score of 7.27 on the COCO dataset, without ever training on COCO, and human raters find Imagen samples to be on par with the … (excerpt)
res_554f506eOur zero-shot FID score is also better than state-of-the-art models trained on COCO, e.g., Make-A-Scene [22] (at 7.6).
clm_13419719We introduce DrawBench, a new structured suite of text prompts for text-to-image evaluation.
res_88ea6b9eWe achieve a new state-of-the-art COCO FID of 7.27.
clm_cef3494cWe introduce DrawBench, a new comprehensive and challenging evaluation benchmark for the text-to-image task.
res_d1aa9a94On DrawBench human evaluation, we find Imagen to outperform all other work, including the concurrent work of DALL-E 2 [54].
res_76e9ddceWe corroborate the results of recent text-guided diffusion work [16, 41, 54] and find that increasing the classifier-free guidance weight improves image-text alignment, but damages image fidelity producing highly saturated … (excerpt)
clm_9dc6eb58Dynamic thresholding: We introduce a new dynamic thresholding method: at each sampling step we set s to a certain percentile absolute pixel value in ˆxt0, and if s > 1, … (excerpt)
res_f83506ecFurthermore, we found Layer Normalization [ 2] for text embeddings in the attention and pooling layers to help considerably improve performance.
res_3258866aWe make several modifications to this U-Net model for improving memory efficiency, inference time and convergence speed (our variant is 2-3x faster in steps/second over the U-Net used in [40, … (excerpt)
res_06311aceFor photorealism, Imagen achieves 39.2% preference rate indicating high image quality generation.
res_7d1d2891Scaling text encoder size is more important than U-Net size.While scaling the size of the diffusion model U-Net improves sample quality, we found scaling the text encoder size to be … (excerpt)
res_75728e61We show that dynamic thresholding results in samples with significantly better photorealism and alignment with text, over static or no thresholding, especially under the presence of large classifier-free guidance weights … (excerpt)
clm_3f453cd7We show that training the super-resolution models with noise conditioning augmentation leads to better CLIP and FID scores.
clm_378a3c93We observe that conditioning over the sequence of text embeddings with cross attention significantly outperforms simple mean or attention based pooling in both sample fidelity as well as image-text alignment.
clm_9baedd9fB.1 Efficient U-Net We introduce a new architectural variant, which we term Efficient U-Net, for our super-resolution models.
res_8b1fa891We make several key modifications to the U-Net architecture, such as shifting of model parameters from high resolution blocks to low resolution, scaling the skip connections by 1/√2 similar to … (excerpt)
res_06d798d3Since lower resolution blocks typically have many more channels, this allows us to increase the model capacity through more model parameters, without egregious memory and computation costs. • When using … (excerpt)
res_51d2d143We observe that scaling the size of the language model text encoders generally results in better image-text 5https://github.com/openai/CLIP/blob/main/model-card.md 21 Category Description Examples Colors Ability to generate objects “A blue colored … (excerpt)
res_cd7e9e51Furthermore, for all values of aug_level, we observe that FID improves considerably with increasing guidance weight upto around 7− 10.
res_646e2c6fAs we scale from 300M parameters to 2B parameters for the U-Net model, we obtain better trade-off curves with increasing model capacity.
res_e67c3eb4Imagen outperforms GLIDE on 8 out of 11 categories on image-text alignment, and 10 out of 11 categories on image fidelity.
Record your result in the Mearc app on your own machine, then use Publish this result with this paper's id (2205.11487) and the claim id shown next to the claim. Mearc creates a file that holds only what you choose to publish. Add it to the public claim data by pull request; a maintainer other than you reviews it before it counts.