ArXiv -> HF Papers

docs/source/en/_toctree.yml

@@ -636,6 +636,8 @@
         title: HunyuanVideo
       - local: api/pipelines/i2vgenxl
         title: I2VGen-XL
+      - local: api/pipelines/kandinsky5_video
+        title: Kandinsky 5.0 Video
       - local: api/pipelines/latte
         title: Latte
       - local: api/pipelines/ltx_video
@@ -654,8 +656,6 @@
         title: Text2Video-Zero
       - local: api/pipelines/wan
         title: Wan
-      - local: api/pipelines/kandinsky5_video
-        title: Kandinsky 5.0 Video
       title: Video
     title: Pipelines
   - sections:

examples/community/README.md

@@ -5488,7 +5488,7 @@ Editing at Scale", many thanks to their contribution!
 
 This implementation of Flux Kontext allows users to pass multiple reference images. Each image is encoded separately, and the resulting latent vectors are concatenated.
 
-As explained in Section 3 of [the paper](https://arxiv.org/pdf/2506.15742), the model's sequence concatenation mechanism can extend its capabilities to handle multiple reference images. However, note that the current version of Flux Kontext was not trained for this use case. In practice, stacking along the first axis does not yield correct results, while stacking along the other two axes appears to work.
+As explained in Section 3 of [the paper](https://huggingface.co/papers/2506.15742), the model's sequence concatenation mechanism can extend its capabilities to handle multiple reference images. However, note that the current version of Flux Kontext was not trained for this use case. In practice, stacking along the first axis does not yield correct results, while stacking along the other two axes appears to work.
 
 ## Example Usage
 

examples/community/regional_prompting_stable_diffusion.py

@@ -490,7 +490,7 @@ def hook_forwards(root_module: torch.nn.Module):
     def prepare_extra_step_kwargs(self, generator, eta):
         # prepare extra kwargs for the scheduler step, since not all schedulers have the same signature
         # eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers.
-        # eta corresponds to η in DDIM paper: https://arxiv.org/abs/2010.02502
+        # eta corresponds to η in DDIM paper: https://huggingface.co/papers/2010.02502
         # and should be between [0, 1]
 
         accepts_eta = "eta" in set(inspect.signature(self.scheduler.step).parameters.keys())
@@ -841,7 +841,7 @@ def stable_diffusion_call(
             num_images_per_prompt (`int`, *optional*, defaults to 1):
                 The number of images to generate per prompt.
             eta (`float`, *optional*, defaults to 0.0):
-                Corresponds to parameter eta (η) from the [DDIM](https://arxiv.org/abs/2010.02502) paper. Only applies
+                Corresponds to parameter eta (η) from the [DDIM](https://huggingface.co/papers/2010.02502) paper. Only applies
                 to the [`~schedulers.DDIMScheduler`], and is ignored in other schedulers.
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
                 A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
@@ -872,7 +872,7 @@ def stable_diffusion_call(
                 [`self.processor`](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
             guidance_rescale (`float`, *optional*, defaults to 0.0):
                 Guidance rescale factor from [Common Diffusion Noise Schedules and Sample Steps are
-                Flawed](https://arxiv.org/pdf/2305.08891.pdf). Guidance rescale factor should fix overexposure when
+                Flawed](https://huggingface.co/papers/2305.08891). Guidance rescale factor should fix overexposure when
                 using zero terminal SNR.
             clip_skip (`int`, *optional*):
                 Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
@@ -1062,7 +1062,7 @@ def stable_diffusion_call(
                     noise_pred = noise_pred_uncond + self.guidance_scale * (noise_pred_text - noise_pred_uncond)
 
                 if self.do_classifier_free_guidance and self.guidance_rescale > 0.0:
-                    # Based on 3.4. in https://arxiv.org/pdf/2305.08891.pdf
+                    # Based on 3.4. in https://huggingface.co/papers/2305.08891
                     noise_pred = rescale_noise_cfg(noise_pred, noise_pred_text, guidance_rescale=self.guidance_rescale)
 
                 # compute the previous noisy sample x_t -> x_t-1
@@ -1668,7 +1668,7 @@ def rescale_noise_cfg(noise_cfg, noise_pred_text, guidance_rescale=0.0):
     r"""
     Rescales `noise_cfg` tensor based on `guidance_rescale` to improve image quality and fix overexposure. Based on
     Section 3.4 from [Common Diffusion Noise Schedules and Sample Steps are
-    Flawed](https://arxiv.org/pdf/2305.08891.pdf).
+    Flawed](https://huggingface.co/papers/2305.08891).
 
     Args:
         noise_cfg (`torch.Tensor`):

src/diffusers/guiders/guider_utils.py

@@ -323,7 +323,7 @@ def rescale_noise_cfg(noise_cfg, noise_pred_text, guidance_rescale=0.0):
     r"""
     Rescales `noise_cfg` tensor based on `guidance_rescale` to improve image quality and fix overexposure. Based on
     Section 3.4 from [Common Diffusion Noise Schedules and Sample Steps are
-    Flawed](https://arxiv.org/pdf/2305.08891.pdf).
+    Flawed](https://huggingface.co/papers/2305.08891).
 
     Args:
         noise_cfg (`torch.Tensor`):

src/diffusers/models/autoencoders/autoencoder_kl_qwenimage.py

@@ -16,7 +16,7 @@
 # QwenImageVAE is further fine-tuned from the Wan Video VAE to achieve improved performance.
 # For more information about the Wan VAE, please refer to:
 # - GitHub: https://github.com/Wan-Video/Wan2.1
-# - arXiv: https://arxiv.org/abs/2503.20314
+# - Paper: https://huggingface.co/papers/2503.20314
 
 from typing import List, Optional, Tuple, Union
 

src/diffusers/pipelines/bria/pipeline_bria.py

@@ -245,7 +245,7 @@ def guidance_scale(self):
         return self._guidance_scale
 
     # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
-    # of the Imagen paper: https://arxiv.org/pdf/2205.11487.pdf . `guidance_scale = 1`
+    # of the Imagen paper: https://huggingface.co/papers/2205.11487 . `guidance_scale = 1`
     # corresponds to doing no classifier free guidance.
     @property
     def do_classifier_free_guidance(self):
@@ -489,11 +489,11 @@ def __call__(
                 in their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is
                 passed will be used. Must be in descending order.
             guidance_scale (`float`, *optional*, defaults to 5.0):
-                Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
-                `guidance_scale` is defined as `w` of equation 2. of [Imagen
-                Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >
-                1`. Higher guidance scale encourages to generate images that are closely linked to the text `prompt`,
-                usually at the expense of lower image quality.
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
             negative_prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is

src/diffusers/pipelines/bria_fibo/pipeline_bria_fibo.py

@@ -337,7 +337,7 @@ def guidance_scale(self):
         return self._guidance_scale
 
     # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
-    # of the Imagen paper: https://arxiv.org/pdf/2205.11487.pdf . `guidance_scale = 1`
+    # of the Imagen paper: https://huggingface.co/papers/2205.11487 . `guidance_scale = 1`
     # corresponds to doing no classifier free guidance.
 
     @property
@@ -498,11 +498,11 @@ def __call__(
                 in their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is
                 passed will be used. Must be in descending order.
             guidance_scale (`float`, *optional*, defaults to 5.0):
-                Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
-                `guidance_scale` is defined as `w` of equation 2. of [Imagen
-                Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >
-                1`. Higher guidance scale encourages to generate images that are closely linked to the text `prompt`,
-                usually at the expense of lower image quality.
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
             negative_prompt (`str` or `List[str]`, *optional*):
                 The prompt or prompts not to guide the image generation. If not defined, one has to pass
                 `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is

src/diffusers/pipelines/ltx/pipeline_ltx.py

@@ -590,9 +590,10 @@ def __call__(
                 the text `prompt`, usually at the expense of lower image quality.
             guidance_rescale (`float`, *optional*, defaults to 0.0):
                 Guidance rescale factor proposed by [Common Diffusion Noise Schedules and Sample Steps are
-                Flawed](https://arxiv.org/pdf/2305.08891.pdf) `guidance_scale` is defined as `φ` in equation 16. of
-                [Common Diffusion Noise Schedules and Sample Steps are Flawed](https://arxiv.org/pdf/2305.08891.pdf).
-                Guidance rescale factor should fix overexposure when using zero terminal SNR.
+                Flawed](https://huggingface.co/papers/2305.08891) `guidance_scale` is defined as `φ` in equation 16. of
+                [Common Diffusion Noise Schedules and Sample Steps are
+                Flawed](https://huggingface.co/papers/2305.08891). Guidance rescale factor should fix overexposure when
+                using zero terminal SNR.
             num_videos_per_prompt (`int`, *optional*, defaults to 1):
                 The number of videos to generate per prompt.
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
@@ -777,7 +778,7 @@ def __call__(
                     noise_pred = noise_pred_uncond + self.guidance_scale * (noise_pred_text - noise_pred_uncond)
 
                     if self.guidance_rescale > 0:
-                        # Based on 3.4. in https://arxiv.org/pdf/2305.08891.pdf
+                        # Based on 3.4. in https://huggingface.co/papers/2305.08891
                         noise_pred = rescale_noise_cfg(
                             noise_pred, noise_pred_text, guidance_rescale=self.guidance_rescale
                         )

src/diffusers/pipelines/ltx/pipeline_ltx_condition.py

@@ -927,9 +927,10 @@ def __call__(
                 the text `prompt`, usually at the expense of lower image quality.
             guidance_rescale (`float`, *optional*, defaults to 0.0):
                 Guidance rescale factor proposed by [Common Diffusion Noise Schedules and Sample Steps are
-                Flawed](https://arxiv.org/pdf/2305.08891.pdf) `guidance_scale` is defined as `φ` in equation 16. of
-                [Common Diffusion Noise Schedules and Sample Steps are Flawed](https://arxiv.org/pdf/2305.08891.pdf).
-                Guidance rescale factor should fix overexposure when using zero terminal SNR.
+                Flawed](https://huggingface.co/papers/2305.08891) `guidance_scale` is defined as `φ` in equation 16. of
+                [Common Diffusion Noise Schedules and Sample Steps are
+                Flawed](https://huggingface.co/papers/2305.08891). Guidance rescale factor should fix overexposure when
+                using zero terminal SNR.
             num_videos_per_prompt (`int`, *optional*, defaults to 1):
                 The number of videos to generate per prompt.
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
@@ -1194,7 +1195,7 @@ def __call__(
                     timestep, _ = timestep.chunk(2)
 
                     if self.guidance_rescale > 0:
-                        # Based on 3.4. in https://arxiv.org/pdf/2305.08891.pdf
+                        # Based on 3.4. in https://huggingface.co/papers/2305.08891
                         noise_pred = rescale_noise_cfg(
                             noise_pred, noise_pred_text, guidance_rescale=self.guidance_rescale
                         )

src/diffusers/pipelines/ltx/pipeline_ltx_image2video.py

@@ -654,9 +654,10 @@ def __call__(
                 the text `prompt`, usually at the expense of lower image quality.
             guidance_rescale (`float`, *optional*, defaults to 0.0):
                 Guidance rescale factor proposed by [Common Diffusion Noise Schedules and Sample Steps are
-                Flawed](https://arxiv.org/pdf/2305.08891.pdf) `guidance_scale` is defined as `φ` in equation 16. of
-                [Common Diffusion Noise Schedules and Sample Steps are Flawed](https://arxiv.org/pdf/2305.08891.pdf).
-                Guidance rescale factor should fix overexposure when using zero terminal SNR.
+                Flawed](https://huggingface.co/papers/2305.08891) `guidance_scale` is defined as `φ` in equation 16. of
+                [Common Diffusion Noise Schedules and Sample Steps are
+                Flawed](https://huggingface.co/papers/2305.08891). Guidance rescale factor should fix overexposure when
+                using zero terminal SNR.
             num_videos_per_prompt (`int`, *optional*, defaults to 1):
                 The number of videos to generate per prompt.
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
@@ -851,7 +852,7 @@ def __call__(
                     timestep, _ = timestep.chunk(2)
 
                     if self.guidance_rescale > 0:
-                        # Based on 3.4. in https://arxiv.org/pdf/2305.08891.pdf
+                        # Based on 3.4. in https://huggingface.co/papers/2305.08891
                         noise_pred = rescale_noise_cfg(
                             noise_pred, noise_pred_text, guidance_rescale=self.guidance_rescale
                         )

src/diffusers/pipelines/prx/pipeline_prx.py

@@ -536,11 +536,11 @@ def __call__(
                 in their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is
                 passed will be used. Must be in descending order.
             guidance_scale (`float`, *optional*, defaults to 4.0):
-                Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
-                `guidance_scale` is defined as `w` of equation 2. of [Imagen
-                Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >
-                1`. Higher guidance scale encourages to generate images that are closely linked to the text `prompt`,
-                usually at the expense of lower image quality.
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
             num_images_per_prompt (`int`, *optional*, defaults to 1):
                 The number of images to generate per prompt.
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):

src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2.py

@@ -415,11 +415,11 @@ def __call__(
                 The number of denoising steps. More denoising steps usually lead to a higher quality image at the
                 expense of slower inference.
             guidance_scale (`float`, defaults to `6.0`):
-                Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
-                `guidance_scale` is defined as `w` of equation 2. of [Imagen
-                Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >
-                1`. Higher guidance scale encourages to generate images that are closely linked to the text `prompt`,
-                usually at the expense of lower image quality.
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality.
             num_videos_per_prompt (`int`, *optional*, defaults to 1):
                 The number of images to generate per prompt.
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):

src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2_diffusion_forcing.py

@@ -647,11 +647,11 @@ def __call__(
                 The number of denoising steps. More denoising steps usually lead to a higher quality image at the
                 expense of slower inference.
             guidance_scale (`float`, defaults to `6.0`):
-                Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
-                `guidance_scale` is defined as `w` of equation 2. of [Imagen
-                Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >
-                1`. Higher guidance scale encourages to generate images that are closely linked to the text `prompt`,
-                usually at the expense of lower image quality. (**6.0 for T2V**, **5.0 for I2V**)
+                Guidance scale as defined in [Classifier-Free Diffusion
+                Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2.
+                of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting
+                `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to
+                the text `prompt`, usually at the expense of lower image quality. (**6.0 for T2V**, **5.0 for I2V**)
             num_videos_per_prompt (`int`, *optional*, defaults to 1):
                 The number of images to generate per prompt.
             generator (`torch.Generator` or `List[torch.Generator]`, *optional*):