正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Diffusion clicked for me when I read about score-based models, a line of work pioneered by Stefano Ermon (et al.) at Stanford. So it was a full-circle moment to collab with him and Inception on a video about training & sampling techniques for making diffusion LLMs faster.

Julia Turc

23,900 subscribers

27,186 次观看 • 5 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

I finally released my new video on YouTube about Diffusion Models / Score-Based Generative Models. Literally planned this for a year and put so much work in. I think this approach to diffusion models is so intuitive and highly recommend giving that a go! Video is 38min long, so you will need some time to watch that haha.

I finally released my new video on YouTube about Diffusion Models / Score-Based Generative Models. Literally planned this for a year and put so much work in. I think this approach to diffusion models is so intuitive and highly recommend giving that a go! Video is 38min long, so you will need some time to watch that haha.

dome | Outlier

54,581 次观看 • 1 年前

New episode of the Information Bottleneck! We talked with Stefano Ermon about why he thinks diffusion LLMs will replace autoregressive ones. Stefano co-invented DDIM, FlashAttention, DPO, and score-based diffusion models. He's a Stanford professor and now runs Inception AI, where they built Mercury II. We go deep but also cover the bigger picture - the startup journey, PhD vs industry, and where AI is heading. A few things that stuck with me: - He thinks of autoregressive models as typewriters and diffusion models as editors. One goes left to right. The other starts messy and refines. - Mercury II (their text difussion model) already beats the fastest autoregressive models on latency-critical stuff as voice agents, code suggestions, anything where you have a tight time budget. And it does it because diffusion generates tokens in parallel instead of one at a time. - We also got into whether AI will actually replace software engineers (his answer: no), PhD vs industry advice, and what it was like going from an ICML best paper to raising money.

New episode of the Information Bottleneck! We talked with Stefano Ermon about why he thinks diffusion LLMs will replace autoregressive ones. Stefano co-invented DDIM, FlashAttention, DPO, and score-based diffusion models. He's a Stanford professor and now runs Inception AI, where they built Mercury II. We go deep but also cover the bigger picture - the startup journey, PhD vs industry, and where AI is heading. A few things that stuck with me: - He thinks of autoregressive models as typewriters and diffusion models as editors. One goes left to right. The other starts messy and refines. - Mercury II (their text difussion model) already beats the fastest autoregressive models on latency-critical stuff as voice agents, code suggestions, anything where you have a tight time budget. And it does it because diffusion generates tokens in parallel instead of one at a time. - We also got into whether AI will actually replace software engineers (his answer: no), PhD vs industry advice, and what it was like going from an ICML best paper to raising money.

Ravid Shwartz Ziv

22,350 次观看 • 4 个月前

'Rerender A Video' by Shuai Yang et al. uses a cluster of techniques to remove the flickering and temporal inconsistency that happens when using stable diffusion for video, while still being compatible with LoRA/ControlNet-driven content.

'Rerender A Video' by Shuai Yang et al. uses a cluster of techniques to remove the flickering and temporal inconsistency that happens when using stable diffusion for video, while still being compatible with LoRA/ControlNet-driven content.

Ben Ferns

726,935 次观看 • 3 年前

What if language models didn't have to generate one token at a time? Our CEO Stefano Ermon joined @TBPN to break down how Mercury 2's diffusion LLM hits 1,000+ tok/s on standard NVIDIA GPUs — and why speed changes the product for coding, voice agents, and search.

What if language models didn't have to generate one token at a time? Our CEO Stefano Ermon joined @TBPN to break down how Mercury 2's diffusion LLM hits 1,000+ tok/s on standard NVIDIA GPUs — and why speed changes the product for coding, voice agents, and search.

Inception

27,108 次观看 • 5 个月前

LLM-grounded Diffusion: Enhancing Prompt Understanding of Text-to-Image Diffusion Models with Large Language Models paper page: github: Recent advancements in text-to-image generation with diffusion models have yielded remarkable results synthesizing highly realistic and diverse images. However, these models still encounter difficulties when generating images from prompts that demand spatial or common sense reasoning. We propose to equip diffusion models with enhanced reasoning capabilities by using off-the-shelf pretrained large language models (LLMs) in a novel two-stage generation process. First, we adapt an LLM to be a text-guided layout generator through in-context learning. When provided with an image prompt, an LLM outputs a scene layout in the form of bounding boxes along with corresponding individual descriptions. Second, we steer a diffusion model with a novel controller to generate images conditioned on the layout. Both stages utilize frozen pretrained models without any LLM or diffusion model parameter optimization. We validate the superiority of our design by demonstrating its ability to outperform the base diffusion model in accurately generating images according to prompts that necessitate both language and spatial reasoning. Additionally, our method naturally allows dialog-based scene specification and is able to handle prompts in a language that is not well-supported by the underlying diffusion model.

LLM-grounded Diffusion: Enhancing Prompt Understanding of Text-to-Image Diffusion Models with Large Language Models paper page: github: Recent advancements in text-to-image generation with diffusion models have yielded remarkable results synthesizing highly realistic and diverse images. However, these models still encounter difficulties when generating images from prompts that demand spatial or common sense reasoning. We propose to equip diffusion models with enhanced reasoning capabilities by using off-the-shelf pretrained large language models (LLMs) in a novel two-stage generation process. First, we adapt an LLM to be a text-guided layout generator through in-context learning. When provided with an image prompt, an LLM outputs a scene layout in the form of bounding boxes along with corresponding individual descriptions. Second, we steer a diffusion model with a novel controller to generate images conditioned on the layout. Both stages utilize frozen pretrained models without any LLM or diffusion model parameter optimization. We validate the superiority of our design by demonstrating its ability to outperform the base diffusion model in accurately generating images according to prompts that necessitate both language and spatial reasoning. Additionally, our method naturally allows dialog-based scene specification and is able to handle prompts in a language that is not well-supported by the underlying diffusion model.

AK

83,657 次观看 • 3 年前

Wow, we can steer diffusion models at inference time! Introducing Diffusion Tree Sampling (DTS): a search-based approach inspired by Monte Carlo Tree Search that turns inference into an anytime, reward-guided optimization process. Diffusion Tree Sampling (DTS) produces asymptotically exact samples from the target distribution in the limit of infinite rollouts, and its greedy variant, Diffusion Tree Search (DTS⋆), performs a global search for high reward samples. The results are pretty impressive: - On MNIST and CIFAR-10 class-conditional generation, DTS matches the FID of the best-performing baseline with up to 10× less compute. - In text-to-image generation and language completion tasks, DTS⋆ effectively searches for high reward samples that match best-of-N with up to 5× less compute.

Wow, we can steer diffusion models at inference time! Introducing Diffusion Tree Sampling (DTS): a search-based approach inspired by Monte Carlo Tree Search that turns inference into an anytime, reward-guided optimization process. Diffusion Tree Sampling (DTS) produces asymptotically exact samples from the target distribution in the limit of infinite rollouts, and its greedy variant, Diffusion Tree Search (DTS⋆), performs a global search for high reward samples. The results are pretty impressive: - On MNIST and CIFAR-10 class-conditional generation, DTS matches the FID of the best-performing baseline with up to 10× less compute. - In text-to-image generation and language completion tasks, DTS⋆ effectively searches for high reward samples that match best-of-N with up to 5× less compute.

机器之心 JIQIZHIXIN

19,037 次观看 • 1 年前

Speed and quality can finally coexist in diffusion-based language generation. Introducing DiDi-Instruct, a Discrete Diffusion Divergence Instruct method that distills a pre-trained discrete diffusion language model (dLLM) into a few-step student for ultra-fast generation. Built on integral KL-divergence minimization, DiDi-Instruct achieves up to 64× faster decoding, surpasses both its teacher and GPT-2, and cuts training time by 20×. Ultra-Fast Language Generation via Discrete Diffusion Divergence Instruct Paper: Code: Project: Our report: 📬 #PapersAccepted by Jiqizhixin

Speed and quality can finally coexist in diffusion-based language generation. Introducing DiDi-Instruct, a Discrete Diffusion Divergence Instruct method that distills a pre-trained discrete diffusion language model (dLLM) into a few-step student for ultra-fast generation. Built on integral KL-divergence minimization, DiDi-Instruct achieves up to 64× faster decoding, surpasses both its teacher and GPT-2, and cuts training time by 20×. Ultra-Fast Language Generation via Discrete Diffusion Divergence Instruct Paper: Code: Project: Our report: 📬 #PapersAccepted by Jiqizhixin

机器之心 JIQIZHIXIN

18,174 次观看 • 9 个月前

TokenFlow: Consistent Diffusion Features for Consistent Video Editing paper page: The generative AI revolution has recently expanded to videos. Nevertheless, current state-of-the-art video models are still lagging behind image models in terms of visual quality and user control over the generated content. In this work, we present a framework that harnesses the power of a text-to-image diffusion model for the task of text-driven video editing. Specifically, given a source video and a target text-prompt, our method generates a high-quality video that adheres to the target text, while preserving the spatial layout and motion of the input video. Our method is based on a key observation that consistency in the edited video can be obtained by enforcing consistency in the diffusion feature space. We achieve this by explicitly propagating diffusion features based on inter-frame correspondences, readily available in the model. Thus, our framework does not require any training or fine-tuning, and can work in conjunction with any off-the-shelf text-to-image editing method. We demonstrate state-of-the-art editing results on a variety of real-world videos.

TokenFlow: Consistent Diffusion Features for Consistent Video Editing paper page: The generative AI revolution has recently expanded to videos. Nevertheless, current state-of-the-art video models are still lagging behind image models in terms of visual quality and user control over the generated content. In this work, we present a framework that harnesses the power of a text-to-image diffusion model for the task of text-driven video editing. Specifically, given a source video and a target text-prompt, our method generates a high-quality video that adheres to the target text, while preserving the spatial layout and motion of the input video. Our method is based on a key observation that consistency in the edited video can be obtained by enforcing consistency in the diffusion feature space. We achieve this by explicitly propagating diffusion features based on inter-frame correspondences, readily available in the model. Thus, our framework does not require any training or fine-tuning, and can work in conjunction with any off-the-shelf text-to-image editing method. We demonstrate state-of-the-art editing results on a variety of real-world videos.

AK

143,469 次观看 • 3 年前

MVDream: Multi-view Diffusion for 3D Generation paper page: propose MVDream, a multi-view diffusion model that is able to generate geometrically consistent multi-view images from a given text prompt. By leveraging image diffusion models pre-trained on large-scale web datasets and a multi-view dataset rendered from 3D assets, the resulting multi-view diffusion model can achieve both the generalizability of 2D diffusion and the consistency of 3D data. Such a model can thus be applied as a multi-view prior for 3D generation via Score Distillation Sampling, where it greatly improves the stability of existing 2D-lifting methods by solving the 3D consistency problem. Finally, we show that the multi-view diffusion model can also be fine-tuned under a few shot setting for personalized 3D generation, i.e. DreamBooth3D application, where the consistency can be maintained after learning the subject identity.

MVDream: Multi-view Diffusion for 3D Generation paper page: propose MVDream, a multi-view diffusion model that is able to generate geometrically consistent multi-view images from a given text prompt. By leveraging image diffusion models pre-trained on large-scale web datasets and a multi-view dataset rendered from 3D assets, the resulting multi-view diffusion model can achieve both the generalizability of 2D diffusion and the consistency of 3D data. Such a model can thus be applied as a multi-view prior for 3D generation via Score Distillation Sampling, where it greatly improves the stability of existing 2D-lifting methods by solving the 3D consistency problem. Finally, we show that the multi-view diffusion model can also be fine-tuned under a few shot setting for personalized 3D generation, i.e. DreamBooth3D application, where the consistency can be maintained after learning the subject identity.

AK

294,496 次观看 • 2 年前

Presto! Distilling Steps and Layers for Accelerating Music Generation Despite advances in diffusion-based text-to-music (TTM) methods, efficient, high-quality generation remains a challenge. We introduce Presto!, an approach to inference acceleration for score-based diffusion transformers via reducing both sampling steps and cost per step. To reduce steps, we develop a new score-based distribution matching distillation (DMD) method for the EDM-family of diffusion models, the first GAN-based distillation method for TTM. To reduce the cost per step, we develop a simple, but powerful improvement to a recent layer distillation method that improves learning via better preserving hidden state variance. Finally, we combine our step and layer distillation methods together for a dual-faceted approach. We evaluate our step and layer distillation methods independently and show each yield best-in-class performance. Our combined distillation method can generate high-quality outputs with improved diversity, accelerating our base model by 10-18x (230/435ms latency for 32 second mono/stereo 44.1kHz, 15x faster than comparable SOTA) -- the fastest high-quality TTM to our knowledge.

Presto! Distilling Steps and Layers for Accelerating Music Generation Despite advances in diffusion-based text-to-music (TTM) methods, efficient, high-quality generation remains a challenge. We introduce Presto!, an approach to inference acceleration for score-based diffusion transformers via reducing both sampling steps and cost per step. To reduce steps, we develop a new score-based distribution matching distillation (DMD) method for the EDM-family of diffusion models, the first GAN-based distillation method for TTM. To reduce the cost per step, we develop a simple, but powerful improvement to a recent layer distillation method that improves learning via better preserving hidden state variance. Finally, we combine our step and layer distillation methods together for a dual-faceted approach. We evaluate our step and layer distillation methods independently and show each yield best-in-class performance. Our combined distillation method can generate high-quality outputs with improved diversity, accelerating our base model by 10-18x (230/435ms latency for 32 second mono/stereo 44.1kHz, 15x faster than comparable SOTA) -- the fastest high-quality TTM to our knowledge.

AK

30,430 次观看 • 1 年前

Been thinking about what this paper really means. "Video diffusion" and "World Models" are becoming synonymous. Neural Computers are basically video diffusion world models for terminal envs and GUI. Lots of talk last week about automating Manim videos. In theory, we should be able to train these world models on a 10000 hours of diverse manim videos and "see where it goes" If a NC can generate outputs to terminal commands, it should be able to generate videos like this directly from prompt too. Without writing code.

Been thinking about what this paper really means. "Video diffusion" and "World Models" are becoming synonymous. Neural Computers are basically video diffusion world models for terminal envs and GUI. Lots of talk last week about automating Manim videos. In theory, we should be able to train these world models on a 10000 hours of diverse manim videos and "see where it goes" If a NC can generate outputs to terminal commands, it should be able to generate videos like this directly from prompt too. Without writing code.

AVB

119,949 次观看 • 3 个月前

Inception's founding team came together at Stanford. Aditya Grover was Stefano Ermon's second PhD student. Volodymyr Kuleshov 🇺🇦 joined the group as a postdoc. Aditya and Volo shared an office. Years later, Stefano's lab hit a breakthrough on diffusion for language. Volo's group at Cornell was publishing adjacent work. Aditya's research at UCLA overlapped with the direction. Part 3 of our founder story series with Tim Tully at Menlo Ventures ↓

Inception's founding team came together at Stanford. Aditya Grover was Stefano Ermon's second PhD student. Volodymyr Kuleshov 🇺🇦 joined the group as a postdoc. Aditya and Volo shared an office. Years later, Stefano's lab hit a breakthrough on diffusion for language. Volo's group at Cornell was publishing adjacent work. Aditya's research at UCLA overlapped with the direction. Part 3 of our founder story series with Tim Tully at Menlo Ventures ↓

Inception

16,653 次观看 • 3 个月前

I love when Grace stops by rehearsals and we can teach her some new dance moves. I’m so lucky to be her uncle. Makes me think about when I was a little kid watching my sister in rehearsals on her season of DWTS, it really is a full circle moment ❤️ Dancing with the Stars #DWTS

I love when Grace stops by rehearsals and we can teach her some new dance moves. I’m so lucky to be her uncle. Makes me think about when I was a little kid watching my sister in rehearsals on her season of DWTS, it really is a full circle moment ❤️ Dancing with the Stars #DWTS

Robert Irwin

333,693 次观看 • 8 个月前

Since we're going full circle today, I thought it was so cute and funny that Jungkook used to brag to Jimin about training in the US for a month and Jimin seemed so impressed 😂

Since we're going full circle today, I thought it was so cute and funny that Jungkook used to brag to Jimin about training in the US for a month and Jimin seemed so impressed 😂

mmmusic⁷⁼¹ BTS is BACK

81,646 次观看 • 7 个月前

Font design is about to get a huge upgrade. Traditionally, given the same font, the letter ‘A’ looks identical everywhere. How about a smart & magical font that draws ‘A’ differently based on the semantic meaning of the word? Smells like Stable Diffusion? Deep dive with me: 🧵

Font design is about to get a huge upgrade. Traditionally, given the same font, the letter ‘A’ looks identical everywhere. How about a smart & magical font that draws ‘A’ differently based on the semantic meaning of the word? Smells like Stable Diffusion? Deep dive with me: 🧵

Jim Fan

3,830,522 次观看 • 3 年前

Reminds me of when I was trying to watch this video about FNAF ransomware, and things were all normal until at this point. Never clicked off a video so fast in my life.

Reminds me of when I was trying to watch this video about FNAF ransomware, and things were all normal until at this point. Never clicked off a video so fast in my life.

SomeFunniThing

103,810 次观看 • 6 个月前

In Prompt Engineering for Vision Models, taught by Abby Jacques Verre and Caleb Kaiser of Comet , you’ll learn how to prompt and fine-tune vision models for personalized image generation, image editing, object detection and segmentation. The prompts you'll use for vision models could be text, point coordinates, or bounding boxes, depending on the model. You'll also learn to tune hyperparameters to shape the output. Models you'll use include Segment-Anything Model (SAM), OWL-ViT, and Stable Diffusion. You'll also learn to fine-tune Stable Diffusion to generate personalized images (say, an image of a specific person), using a handful of images for training. As an example of a multi-step workflow, you'll use OWL-ViT to detect an object based on a text prompt, then pass the bounding box to SAM to create a segmentation mask, and input that mask into Stable Diffusion to replace the original object with a new one based on a text prompt. Controlling vision models can be tricky; this course will teach prompting and fine-tuning techniques to get precise control over their output. Get started here:

In Prompt Engineering for Vision Models, taught by Abby Jacques Verre and Caleb Kaiser of Comet , you’ll learn how to prompt and fine-tune vision models for personalized image generation, image editing, object detection and segmentation. The prompts you'll use for vision models could be text, point coordinates, or bounding boxes, depending on the model. You'll also learn to tune hyperparameters to shape the output. Models you'll use include Segment-Anything Model (SAM), OWL-ViT, and Stable Diffusion. You'll also learn to fine-tune Stable Diffusion to generate personalized images (say, an image of a specific person), using a handful of images for training. As an example of a multi-step workflow, you'll use OWL-ViT to detect an object based on a text prompt, then pass the bounding box to SAM to create a segmentation mask, and input that mask into Stable Diffusion to replace the original object with a new one based on a text prompt. Controlling vision models can be tricky; this course will teach prompting and fine-tuning techniques to get precise control over their output. Get started here:

Andrew Ng

151,198 次观看 • 2 年前

Marty at Id did ask for a video from me, but there was a miscommunication about when he needed it, so I didn’t get it done by the actual anniversary…

Marty at Id did ask for a video from me, but there was a miscommunication about when he needed it, so I didn’t get it done by the actual anniversary…

John Carmack

680,567 次观看 • 6 个月前