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。

🥳 We present our #ICASSP2024 paper: A diffusion model that generates production-quality (bass) audio stems to any audio input. 😎🎸 According to our experience, that's more useful to artists than generating full mixes. 🙃 📜Paper: 🎶Demo: by Marco Pasini 👈💪SonyCSL(Paris)_Music Team #MusicAI

Stefan Lattner

1,614 subscribers

20,284 次观看 • 2 年前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

3 条评论

Xavier Manuel Mountain 的头像

Xavier Manuel Mountain2 年前

Seriously dope! It’s about to be the digital renaissance in music real quick.

SteveRatatata 的头像

SteveRatatata2 年前

Thanks for the great work! Are there any plans to release the code?

m҉a҉r҉k҉ 的头像

m҉a҉r҉k҉2 年前

this is amazing!

相关视频

🥳Excited to share our latest work: "Diff-A-Riff"! 🥁 A Latent Diffusion Model that generates instrumental accompaniments for any musical input, specifically tailored for music producers! It's faster, lighter, and produces superior audio quality. Control via text/audio references. 48kHz sample rate, (pseudo) stereo, ~3Gb memory, takes 6 seconds to generate 90 seconds of music. Trained on a single GPU. 📜 🎶 🎸 "Diff-A-Riff" adapts to any musical input, following the artist's unique style. 🎛️ Optional controls via text prompts, audio references, interpolation slider, pseudo-stereo width and loop intensity. 🎚️ It produces state-of-the-art audio quality indistinguishable from real data by human raters and operates at unprecedented speed. 🧠 "Diff-A-Riff" is smaller and more efficient than previous models thanks to its Consistency Autoencoder, making it accessible and practical for various applications. Big shoutout to my outer space colleagues: Javier Nistal, the Machine in "machine learning" 🚄, Marco Pasini, the neural net whisperer 🤫, Cyran Aouameur, the troubleshootah 🛠️, Maarten Grachten, aka MaartenGPT 🤖. #Teamwork #AI #MusicTech #Innovation Javier Nistal Marco Pasini Cyran Aouameur SonyCSL(Paris)_Music Team

🥳Excited to share our latest work: "Diff-A-Riff"! 🥁 A Latent Diffusion Model that generates instrumental accompaniments for any musical input, specifically tailored for music producers! It's faster, lighter, and produces superior audio quality. Control via text/audio references. 48kHz sample rate, (pseudo) stereo, ~3Gb memory, takes 6 seconds to generate 90 seconds of music. Trained on a single GPU. 📜 🎶 🎸 "Diff-A-Riff" adapts to any musical input, following the artist's unique style. 🎛️ Optional controls via text prompts, audio references, interpolation slider, pseudo-stereo width and loop intensity. 🎚️ It produces state-of-the-art audio quality indistinguishable from real data by human raters and operates at unprecedented speed. 🧠 "Diff-A-Riff" is smaller and more efficient than previous models thanks to its Consistency Autoencoder, making it accessible and practical for various applications. Big shoutout to my outer space colleagues: Javier Nistal, the Machine in "machine learning" 🚄, Marco Pasini, the neural net whisperer 🤫, Cyran Aouameur, the troubleshootah 🛠️, Maarten Grachten, aka MaartenGPT 🤖. #Teamwork #AI #MusicTech #Innovation Javier Nistal Marco Pasini Cyran Aouameur SonyCSL(Paris)_Music Team

Stefan Lattner

17,710 次观看 • 2 年前

The reception since we launched the first unified audio + text + image to video model in production last week has been incredible! We wanted to take a moment and highlight some firsts: - the first omnimodal modal that jointly operates over text, image, and audio. Fun fact: we shipped our first audio-to-video model Character-1 last June! - Character-3 is generating 5 minute uncut clips, and capable of temporally stable longform prediction. - Character-3 handles realistic humans, cartoons, side profiles, dynamic backgrounds, anime, animals, and more. - Character-3 sets a new standard for fast & efficient video diffusion at HD and SD resolutions, leveraging our advancements in realtime video generation. We'll have an announcement soon, but you can read more about an earlier version of this line of work in our paper here:

The reception since we launched the first unified audio + text + image to video model in production last week has been incredible! We wanted to take a moment and highlight some firsts: - the first omnimodal modal that jointly operates over text, image, and audio. Fun fact: we shipped our first audio-to-video model Character-1 last June! - Character-3 is generating 5 minute uncut clips, and capable of temporally stable longform prediction. - Character-3 handles realistic humans, cartoons, side profiles, dynamic backgrounds, anime, animals, and more. - Character-3 sets a new standard for fast & efficient video diffusion at HD and SD resolutions, leveraging our advancements in realtime video generation. We'll have an announcement soon, but you can read more about an earlier version of this line of work in our paper here:

Hedra

180,676 次观看 • 1 年前

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

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

Today we’re launching Stable Audio 2.5: The first audio model built for enterprise-grade sound production 🔊 Audio influences brand engagement by 86%, but few enterprises are leveraging audio as an extension of their brand, making customized sound an untapped differentiator. Stable Audio 2.5 is purpose-built for this opportunity to create customizable, high-quality audio at scale, with capabilities that include: ▶️ Improved musical composition: Generate full songs with multi-part structure, meaning a clear intro, middle, and outro. ▶️ Audio inpainting: Input audio, select where the track should start, and the model uses the context to generate the rest of the track. ▶️ Customization: Our team can fine-tune Stable Audio 2.5 to help enterprises create the right sound for their brand. ▶️ Faster inference: The model can generate up to three-minute long tracks in under two seconds on a GPU, outputting in just eight steps (compared to ~50 in the previous model). You can learn more here 👉

Stability AI

67,618 次观看 • 11 个月前

Excited to share our latest work on 🎧spatial audio-driven human motion generation. We aim to tackle a largely underexplored yet important problem of enabling virtual humans to move naturally in response to spatial audio—capturing not just what is heard, but also where the sound is coming from. To this end, we introduce the Spatial Audio-Driven Human Motion (SAM) dataset—the first comprehensive dataset featuring paired high-quality human motion and spatial audio recordings. For benchmarking, we develop a generative framework for human MOtion generation driven by SPAtial audio, termed MOSPA, which learns to synthesize realistic and diverse human motions conditioned on spatial audio input. We hope this research could provide a foundation for future research in spatial perception, virtual characters, and embodied AI. The dataset and model will be open-sourced soon. A big thank you to our intern, Shuyang Xu, for the wonderful collaboration! Congratulations, Shuyang! Project page: Paper: Video: #Animation #CG #CV #AIGC #DL #Deeplearning #Motion #Graphics #AI #GenerativeAI

Excited to share our latest work on 🎧spatial audio-driven human motion generation. We aim to tackle a largely underexplored yet important problem of enabling virtual humans to move naturally in response to spatial audio—capturing not just what is heard, but also where the sound is coming from. To this end, we introduce the Spatial Audio-Driven Human Motion (SAM) dataset—the first comprehensive dataset featuring paired high-quality human motion and spatial audio recordings. For benchmarking, we develop a generative framework for human MOtion generation driven by SPAtial audio, termed MOSPA, which learns to synthesize realistic and diverse human motions conditioned on spatial audio input. We hope this research could provide a foundation for future research in spatial perception, virtual characters, and embodied AI. The dataset and model will be open-sourced soon. A big thank you to our intern, Shuyang Xu, for the wonderful collaboration! Congratulations, Shuyang! Project page: Paper: Video: #Animation #CG #CV #AIGC #DL #Deeplearning #Motion #Graphics #AI #GenerativeAI

Zhiyang (Frank) Dou

14,610 次观看 • 1 年前

Today we're announcing the open-source release of HunyuanVideo-Foley, our new end-to-end Text-Video-to-Audio (TV2A) framework for generating high-fidelity audio.🚀 This tool empowers creators in video production, filmmaking, and game development to generate professional-grade audio that precisely aligns with visual dynamics and semantic context, addressing key challenges in V2A generation.🔊 Key Innovations: 🔹Exceptional Generalization: Trained on a massive 100k-hour multimodal dataset, the model generates contextually-aware soundscapes for a wide range of scenes, from natural landscapes to animated shorts. 🔹Balanced Multimodal Response: Our innovative multimodal diffusion transformer (MMDiT) architecture ensures the model balances video and text cues, generating rich, layered sound effects that capture every detail—from the main subject to subtle background elements. 🔹High-Fidelity Audio: Using a Representation Alignment (REPA) loss function and a powerful Audio VAE, we've improved generation stability and producing professional-grade audio, free of noise and inconsistencies. HunyuanVideo-Foley achieves SOTA on multiple benchmarks, surpassing all open-source models in audio quality, visual-semantic alignment, and temporal alignment. 👉Try it now: 🌐Project Page: 🔗Code: 📄Technical Report: 🤗Hugging Face:

Today we're announcing the open-source release of HunyuanVideo-Foley, our new end-to-end Text-Video-to-Audio (TV2A) framework for generating high-fidelity audio.🚀 This tool empowers creators in video production, filmmaking, and game development to generate professional-grade audio that precisely aligns with visual dynamics and semantic context, addressing key challenges in V2A generation.🔊 Key Innovations: 🔹Exceptional Generalization: Trained on a massive 100k-hour multimodal dataset, the model generates contextually-aware soundscapes for a wide range of scenes, from natural landscapes to animated shorts. 🔹Balanced Multimodal Response: Our innovative multimodal diffusion transformer (MMDiT) architecture ensures the model balances video and text cues, generating rich, layered sound effects that capture every detail—from the main subject to subtle background elements. 🔹High-Fidelity Audio: Using a Representation Alignment (REPA) loss function and a powerful Audio VAE, we've improved generation stability and producing professional-grade audio, free of noise and inconsistencies. HunyuanVideo-Foley achieves SOTA on multiple benchmarks, surpassing all open-source models in audio quality, visual-semantic alignment, and temporal alignment. 👉Try it now: 🌐Project Page: 🔗Code: 📄Technical Report: 🤗Hugging Face:

Tencent Hy

122,706 次观看 • 11 个月前

Meta announces Movie Gen A Cast of Media Foundation Models We present Movie Gen, a cast of foundation models that generates high-quality, 1080p HD videos with different aspect ratios and synchronized audio. We also show additional capabilities such as precise instruction-based video editing and generation of personalized videos based on a user’s image. Our models set a new state-of-the-art on multiple tasks: text-to-video synthesis, video personalization, video editing, video-to-audio generation, and text-to-audio generation. Our largest video generation model is a 30B parameter transformer trained with a maximum context length of 73K video tokens, corresponding to a generated video of 16 seconds at 16 frames-per-second. We show multiple technical innovations and simplifications on the architecture, latent spaces, training objectives and recipes, data curation, evaluation protocols, parallelization techniques, and inference optimizations that allow us to reap the benefits of scaling pre-training data, model size, and training compute for training large scale media generation models. We hope this paper helps the research community to accelerate progress and innovation in media generation models

Meta announces Movie Gen A Cast of Media Foundation Models We present Movie Gen, a cast of foundation models that generates high-quality, 1080p HD videos with different aspect ratios and synchronized audio. We also show additional capabilities such as precise instruction-based video editing and generation of personalized videos based on a user’s image. Our models set a new state-of-the-art on multiple tasks: text-to-video synthesis, video personalization, video editing, video-to-audio generation, and text-to-audio generation. Our largest video generation model is a 30B parameter transformer trained with a maximum context length of 73K video tokens, corresponding to a generated video of 16 seconds at 16 frames-per-second. We show multiple technical innovations and simplifications on the architecture, latent spaces, training objectives and recipes, data curation, evaluation protocols, parallelization techniques, and inference optimizations that allow us to reap the benefits of scaling pre-training data, model size, and training compute for training large scale media generation models. We hope this paper helps the research community to accelerate progress and innovation in media generation models

AK

62,719 次观看 • 1 年前

VITA Towards Open-Source Interactive Omni Multimodal LLM discuss: The remarkable multimodal capabilities and interactive experience of GPT-4o underscore their necessity in practical applications, yet open-source models rarely excel in both areas. In this paper, we introduce VITA, the first-ever open-source Multimodal Large Language Model (MLLM) adept at simultaneous processing and analysis of Video, Image, Text, and Audio modalities, and meanwhile has an advanced multimodal interactive experience. Starting from Mixtral 8x7B as a language foundation, we expand its Chinese vocabulary followed by bilingual instruction tuning. We further endow the language model with visual and audio capabilities through two-stage multi-task learning of multimodal alignment and instruction tuning. VITA demonstrates robust foundational capabilities of multilingual, vision, and audio understanding, as evidenced by its strong performance across a range of both unimodal and multimodal benchmarks. Beyond foundational capabilities, we have made considerable progress in enhancing the natural multimodal human-computer interaction experience. To the best of our knowledge, we are the first to exploit non-awakening interaction and audio interrupt in MLLM. VITA is the first step for the open-source community to explore the seamless integration of multimodal understanding and interaction. While there is still lots of work to be done on VITA to get close to close-source counterparts, we hope that its role as a pioneer can serve as a cornerstone for subsequent research.

VITA Towards Open-Source Interactive Omni Multimodal LLM discuss: The remarkable multimodal capabilities and interactive experience of GPT-4o underscore their necessity in practical applications, yet open-source models rarely excel in both areas. In this paper, we introduce VITA, the first-ever open-source Multimodal Large Language Model (MLLM) adept at simultaneous processing and analysis of Video, Image, Text, and Audio modalities, and meanwhile has an advanced multimodal interactive experience. Starting from Mixtral 8x7B as a language foundation, we expand its Chinese vocabulary followed by bilingual instruction tuning. We further endow the language model with visual and audio capabilities through two-stage multi-task learning of multimodal alignment and instruction tuning. VITA demonstrates robust foundational capabilities of multilingual, vision, and audio understanding, as evidenced by its strong performance across a range of both unimodal and multimodal benchmarks. Beyond foundational capabilities, we have made considerable progress in enhancing the natural multimodal human-computer interaction experience. To the best of our knowledge, we are the first to exploit non-awakening interaction and audio interrupt in MLLM. VITA is the first step for the open-source community to explore the seamless integration of multimodal understanding and interaction. While there is still lots of work to be done on VITA to get close to close-source counterparts, we hope that its role as a pioneer can serve as a cornerstone for subsequent research.

AK

23,958 次观看 • 2 年前

Congratulations to the research team at our sister company Eyeline on their latest research paper - “Lux Post Facto: Learning Portrait Performance Relighting with Conditional Video Diffusion and a Hybrid Dataset” - which will be presented at #CVPR2025 in Nashville. Lighting plays a crucial role in visual storytelling, shaping the scene, adding depth, and enhancing emotional impact. However, achieving high-quality lighting traditionally requires professional expertise and costly equipment. This new research introduces a method that enables content creators to achieve compelling lighting in portrait videos as a simple post-production process, making high-quality lighting more accessible than ever. A well-deserved recognition for the authors behind this innovation: Yiqun Mei, Mingming He, LiMa, Julien Philip, Wenqi Xian ✈️ CVPR, David George, Xueming Yu, Gabriel Dedic, Ahmet Levent Taşel, Ning Yu Vishal Patel and Paul Debevec . Project Page: Paper: YouTube:

Congratulations to the research team at our sister company Eyeline on their latest research paper - “Lux Post Facto: Learning Portrait Performance Relighting with Conditional Video Diffusion and a Hybrid Dataset” - which will be presented at #CVPR2025 in Nashville. Lighting plays a crucial role in visual storytelling, shaping the scene, adding depth, and enhancing emotional impact. However, achieving high-quality lighting traditionally requires professional expertise and costly equipment. This new research introduces a method that enables content creators to achieve compelling lighting in portrait videos as a simple post-production process, making high-quality lighting more accessible than ever. A well-deserved recognition for the authors behind this innovation: Yiqun Mei, Mingming He, LiMa, Julien Philip, Wenqi Xian ✈️ CVPR, David George, Xueming Yu, Gabriel Dedic, Ahmet Levent Taşel, Ning Yu Vishal Patel and Paul Debevec . Project Page: Paper: YouTube:

Scanline VFX - Powered by Netflix

11,633 次观看 • 1 年前

Modern AI is confined to the digital world. At Skild AI, we are building towards AGI for the real world, unconstrained by robot type or task — a single, omni-bodied brain. Today, we are sharing our journey, starting with early milestones, with more to come in the weeks ahead. Our Mission: Artificial General Intelligence grounded in the physical world. We believe AGI that can truly understand and reason in the real world can only be built through grounding in the physical world. Our Vision: Any robot, Any task, One brain. We tackle robotics in its full generality – building a continually improving, omni-bodied brain that can control any hardware for any task. Who are we? A passionate group of scientists & engineers driven by our shared vision. We have been researching AI and robotics for more than a decade. Our team includes pioneers of self-supervised learning, curiosity-driven exploration, end-to-end sim2real for visual locomotion, dexterous manipulation, learning from human videos, robot parkour, and many more. Many of these works have won awards at top-tier AI and Robotics conferences. Our team has also built production-ready systems at Anduril, Tesla, Nvidia, Meta, Kitty Hawk, Google, Everyday Robotics, and Amazon. Join us in our mission to build the robot brains of tomorrow.

Modern AI is confined to the digital world. At Skild AI, we are building towards AGI for the real world, unconstrained by robot type or task — a single, omni-bodied brain. Today, we are sharing our journey, starting with early milestones, with more to come in the weeks ahead. Our Mission: Artificial General Intelligence grounded in the physical world. We believe AGI that can truly understand and reason in the real world can only be built through grounding in the physical world. Our Vision: Any robot, Any task, One brain. We tackle robotics in its full generality – building a continually improving, omni-bodied brain that can control any hardware for any task. Who are we? A passionate group of scientists & engineers driven by our shared vision. We have been researching AI and robotics for more than a decade. Our team includes pioneers of self-supervised learning, curiosity-driven exploration, end-to-end sim2real for visual locomotion, dexterous manipulation, learning from human videos, robot parkour, and many more. Many of these works have won awards at top-tier AI and Robotics conferences. Our team has also built production-ready systems at Anduril, Tesla, Nvidia, Meta, Kitty Hawk, Google, Everyday Robotics, and Amazon. Join us in our mission to build the robot brains of tomorrow.

Skild AI

383,159 次观看 • 1 年前

Our first short course with Anthropic! Building Towards Computer Use with Anthropic. This teaches you to build an LLM-based agent that uses a computer interface by generating mouse clicks and keystrokes. Computer Use is an important, emerging capability for LLMs that will let AI agents do many more tasks than were possible before, since it lets them interact with interfaces designed for humans to use, rather than only tools that provide explicit API access. I hope you will enjoy learning about it! This course is taught by Anthropic's Head of Curriculum, Colt_Steele. You'll learn to apply image reasoning and tool use to "use" a computer as follows: a model processes an image of the screen, analyzes it to understand what's going on, and navigates the computer via mouse clicks and keystrokes. This course goes through the key building blocks, and culminates in a demo of an AI assistant that uses a web browser to search for a research paper, downloads the PDF, and finally summarizes the paper for you. In detail, you’ll: - Learn about Anthropic's family of models, when to use which one, and make API requests to Claude - Use multi-modal prompts that combine text and image content blocks, and also work with streaming responses - Improve your prompting by using prompt templates, using XML to structure prompts, and providing examples - Implement prompt caching to reduce cost and latency - Apply tool-use to build a chatbot that can call different tools to respond to queries - See all these building blocks come together in Computer Use demo Please sign up here:

Our first short course with Anthropic! Building Towards Computer Use with Anthropic. This teaches you to build an LLM-based agent that uses a computer interface by generating mouse clicks and keystrokes. Computer Use is an important, emerging capability for LLMs that will let AI agents do many more tasks than were possible before, since it lets them interact with interfaces designed for humans to use, rather than only tools that provide explicit API access. I hope you will enjoy learning about it! This course is taught by Anthropic's Head of Curriculum, Colt_Steele. You'll learn to apply image reasoning and tool use to "use" a computer as follows: a model processes an image of the screen, analyzes it to understand what's going on, and navigates the computer via mouse clicks and keystrokes. This course goes through the key building blocks, and culminates in a demo of an AI assistant that uses a web browser to search for a research paper, downloads the PDF, and finally summarizes the paper for you. In detail, you’ll: - Learn about Anthropic's family of models, when to use which one, and make API requests to Claude - Use multi-modal prompts that combine text and image content blocks, and also work with streaming responses - Improve your prompting by using prompt templates, using XML to structure prompts, and providing examples - Implement prompt caching to reduce cost and latency - Apply tool-use to build a chatbot that can call different tools to respond to queries - See all these building blocks come together in Computer Use demo Please sign up here:

Andrew Ng

170,425 次观看 • 1 年前

$*New Paper on AI & Democracy* Imagine two approaches to democracy. The one we have today, where citizens choose a professional politician to represent them and others. Or an augmented form of democracy, where each citizen controls a personalized AI that helps them participate in thousands of nuanced decisions. This second approach is the idea of Augmented Democracy I introduced six years ago at TED. In our latest paper we explore a simplified version of Augmented Democracy by combining off-the-shelf LLMs, such as ChatGPT, with data collected using a collaborative government program builder. This was an online game where people build a personalized government program using proposals extracted from the programs of the candidates of the 2022 presidential election in Brazil. So how accurate are these augmented forms of democracy? Imagine a user who gave us 40 answers. We can use the first 20 to fine-tune a model that we can test using the 20 answers the model didn’t see. We can then compare the accuracy of these predictions with the ones obtained by a “bundle” rule, which assumes that users that self-reported to be from the left or right always chose the proposals from the candidate that shares their political identity. This showed us that LLMs were more accurate at predicting policy preferences than the bundle rule, meaning that the preferences captured in the participation data were more nuanced than a left-right axis, and that the LLMs can capture some of that nuance. Also, the LLMs can choose among policies coming from the same candidate, which is something that we cannot do using a bundle rule. But can these LLMs help us complete the aggregate preferences of the population? Direct or unbundled forms of participation can result in incomplete data when people answer only a fraction of all questions. In our paper, we simulate this incompleteness by sampling the full dataset. We ask how close we can get to the full dataset by using a random sample, or a random sample augmented by predictions made by these LLMs. Overall, we find that LLM-augmented data gets much closer to the full dataset than a pure random sample. These results do not mean that augmented democracy technology is ready, but they means we are in a much better place to continue exploring this idea than six years ago. This paper was a collaborative effort with Jairo Gudino, PhD student at CCL at the University of Toulouse Capitole and Umberto Grandi from IRIT also at the University of Toulouse Capitole. We hope you find these results insightful!$

New Paper on AI & Democracy Imagine two approaches to democracy. The one we have today, where citizens choose a professional politician to represent them and others. Or an augmented form of democracy, where each citizen controls a personalized AI that helps them participate in thousands of nuanced decisions. This second approach is the idea of Augmented Democracy I introduced six years ago at TED. In our latest paper we explore a simplified version of Augmented Democracy by combining off-the-shelf LLMs, such as ChatGPT, with data collected using a collaborative government program builder. This was an online game where people build a personalized government program using proposals extracted from the programs of the candidates of the 2022 presidential election in Brazil. So how accurate are these augmented forms of democracy? Imagine a user who gave us 40 answers. We can use the first 20 to fine-tune a model that we can test using the 20 answers the model didn’t see. We can then compare the accuracy of these predictions with the ones obtained by a “bundle” rule, which assumes that users that self-reported to be from the left or right always chose the proposals from the candidate that shares their political identity. This showed us that LLMs were more accurate at predicting policy preferences than the bundle rule, meaning that the preferences captured in the participation data were more nuanced than a left-right axis, and that the LLMs can capture some of that nuance. Also, the LLMs can choose among policies coming from the same candidate, which is something that we cannot do using a bundle rule. But can these LLMs help us complete the aggregate preferences of the population? Direct or unbundled forms of participation can result in incomplete data when people answer only a fraction of all questions. In our paper, we simulate this incompleteness by sampling the full dataset. We ask how close we can get to the full dataset by using a random sample, or a random sample augmented by predictions made by these LLMs. Overall, we find that LLM-augmented data gets much closer to the full dataset than a pure random sample. These results do not mean that augmented democracy technology is ready, but they means we are in a much better place to continue exploring this idea than six years ago. This paper was a collaborative effort with Jairo Gudino, PhD student at CCL at the University of Toulouse Capitole and Umberto Grandi from IRIT also at the University of Toulouse Capitole. We hope you find these results insightful!

César A. Hidalgo

26,915 次观看 • 1 年前

2 years ago, Riffusion was one of the first music generating AI models, by finetuning Stable Diffusion literally on images of waveforms, and then decoding those as audio again I made a song for my gf with it called "Bunny Trouble, Trouble Bubble" about a year ago, extremely catchy and we loved it But it was capped to 12 seconds and I could only imagine how it'd be as a full song Now Suno is finally good enough and I used it as an input so I could extend it from a 12 second clip to 3 minute song, it's essentially the audio version of img2img It took a few times extending it and half the times the output was very bad but the other half it was really great The quality is still a bit low because Riffusion's quality itself was quite low, so that's not Suno's fault, it just extends the low quality into a full song, what I need is an audio upscaler to fix that The fun thing about this, every year music generating AI will be better and I can try inputting the song again to make it better and better If you wanna do the same with your short clip: 1) go to Suno 2) click Create 3) click Custom 4) Upload Audio, upload your short clip here Your existing clip now gets added to the listing on the right (bit confusing) 5) Now hover over it and tap extend 6) More confusingly it will now only play the extended part not the full song 7) To get the full song, click the 3 vertical dots ... -> Create -> Get Whole Song 8) That will stitch your original input clip and the extended clip into one Now you'll probably have a 1 minute clip, so 9) Extend that WHOLE SONG clip again and do the same stitch thing (Get Whole Song) again Depending on the clips you like and not, choose the ones you wanna continue with by stitching them By now you'll have about a 3 minute song!

2 years ago, Riffusion was one of the first music generating AI models, by finetuning Stable Diffusion literally on images of waveforms, and then decoding those as audio again I made a song for my gf with it called "Bunny Trouble, Trouble Bubble" about a year ago, extremely catchy and we loved it But it was capped to 12 seconds and I could only imagine how it'd be as a full song Now Suno is finally good enough and I used it as an input so I could extend it from a 12 second clip to 3 minute song, it's essentially the audio version of img2img It took a few times extending it and half the times the output was very bad but the other half it was really great The quality is still a bit low because Riffusion's quality itself was quite low, so that's not Suno's fault, it just extends the low quality into a full song, what I need is an audio upscaler to fix that The fun thing about this, every year music generating AI will be better and I can try inputting the song again to make it better and better If you wanna do the same with your short clip: 1) go to Suno 2) click Create 3) click Custom 4) Upload Audio, upload your short clip here Your existing clip now gets added to the listing on the right (bit confusing) 5) Now hover over it and tap extend 6) More confusingly it will now only play the extended part not the full song 7) To get the full song, click the 3 vertical dots ... -> Create -> Get Whole Song 8) That will stitch your original input clip and the extended clip into one Now you'll probably have a 1 minute clip, so 9) Extend that WHOLE SONG clip again and do the same stitch thing (Get Whole Song) again Depending on the clips you like and not, choose the ones you wanna continue with by stitching them By now you'll have about a 3 minute song!

@levelsio

55,113 次观看 • 1 年前

🚀 VoxCPM 2 is live! 🎉 Another open-source AI #TTS model from China — and one that stands shoulder to shoulder with Qwen3-TTS, while bringing everything into a single unified model. After rapid iterations from V1 (zero-shot cloning) to V1.5 (long-form + fine-tuning), #VoxCPM has consistently pushed quality and usability forward. Now, VoxCPM 2 takes it further: 🔹30+ languages — truly global, truly local. 🔹Infinite voice design — type it, hear it, control it. From a whisper to a booming cinematic voice. 🔹Studio-grade audio — 48kHz ultra-high fidelity with emotional depth 🔹Diffusion-Autoregressive cloning — preserves more acoustic and emotional detail than token-based models like Qwen3-TTS 💡 Big shoutout to Grok — used your multi-image video magic for our launch demo. It’s scarily good at keeping visuals consistent across shots. Elon Elon Musk, this one’s for you. 😉 Check the demo & start cloning your dream voice: 🌐 Hugging Face Space: 🤗 Hugging Face Model: 🤖 ModelScope Model: 💻 GitHub： #TTS #AI #VoiceCloning #GrokImagine #ElonMusk #OpenBMB #VoxCPM

🚀 VoxCPM 2 is live! 🎉 Another open-source AI #TTS model from China — and one that stands shoulder to shoulder with Qwen3-TTS, while bringing everything into a single unified model. After rapid iterations from V1 (zero-shot cloning) to V1.5 (long-form + fine-tuning), #VoxCPM has consistently pushed quality and usability forward. Now, VoxCPM 2 takes it further: 🔹30+ languages — truly global, truly local. 🔹Infinite voice design — type it, hear it, control it. From a whisper to a booming cinematic voice. 🔹Studio-grade audio — 48kHz ultra-high fidelity with emotional depth 🔹Diffusion-Autoregressive cloning — preserves more acoustic and emotional detail than token-based models like Qwen3-TTS 💡 Big shoutout to Grok — used your multi-image video magic for our launch demo. It’s scarily good at keeping visuals consistent across shots. Elon Elon Musk, this one’s for you. 😉 Check the demo & start cloning your dream voice: 🌐 Hugging Face Space: 🤗 Hugging Face Model: 🤖 ModelScope Model: 💻 GitHub： #TTS #AI #VoiceCloning #GrokImagine #ElonMusk #OpenBMB #VoxCPM

OpenBMB

559,199 次观看 • 4 个月前

ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillation propose to model the 3D parameter as a random variable instead of a constant as in SDS and present variational score distillation (VSD), a principled particle-based variational framework to explain and address the aforementioned issues in text-to-3D generation. We show that SDS is a special case of VSD and leads to poor samples with both small and large CFG weights. In comparison, VSD works well with various CFG weights as ancestral sampling from diffusion models and simultaneously improves the diversity and sample quality with a common CFG weight (i.e., 7.5). We further present various improvements in the design space for text-to-3D such as distillation time schedule and density initialization, which are orthogonal to the distillation algorithm yet not well explored. Our overall approach, dubbed ProlificDreamer, can generate high rendering resolution (i.e., 512times512) and high-fidelity NeRF with rich structure and complex effects (e.g., smoke and drops). Further, initialized from NeRF, meshes fine-tuned by VSD are meticulously detailed and photo-realistic paper page:

ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillation propose to model the 3D parameter as a random variable instead of a constant as in SDS and present variational score distillation (VSD), a principled particle-based variational framework to explain and address the aforementioned issues in text-to-3D generation. We show that SDS is a special case of VSD and leads to poor samples with both small and large CFG weights. In comparison, VSD works well with various CFG weights as ancestral sampling from diffusion models and simultaneously improves the diversity and sample quality with a common CFG weight (i.e., 7.5). We further present various improvements in the design space for text-to-3D such as distillation time schedule and density initialization, which are orthogonal to the distillation algorithm yet not well explored. Our overall approach, dubbed ProlificDreamer, can generate high rendering resolution (i.e., 512times512) and high-fidelity NeRF with rich structure and complex effects (e.g., smoke and drops). Further, initialized from NeRF, meshes fine-tuned by VSD are meticulously detailed and photo-realistic paper page:

AK

46,151 次观看 • 3 年前

Drivable 3D Gaussian Avatars paper page: present Drivable 3D Gaussian Avatars (D3GA), the first 3D controllable model for human bodies rendered with Gaussian splats. Current photorealistic drivable avatars require either accurate 3D registrations during training, dense input images during testing, or both. The ones based on neural radiance fields also tend to be prohibitively slow for telepresence applications. This work uses the recently presented 3D Gaussian Splatting (3DGS) technique to render realistic humans at real-time framerates, using dense calibrated multi-view videos as input. To deform those primitives, we depart from the commonly used point deformation method of linear blend skinning (LBS) and use a classic volumetric deformation method: cage deformations. Given their smaller size, we drive these deformations with joint angles and keypoints, which are more suitable for communication applications. Our experiments on nine subjects with varied body shapes, clothes, and motions obtain higher-quality results than state-of-the-art methods when using the same training and test data.

Drivable 3D Gaussian Avatars paper page: present Drivable 3D Gaussian Avatars (D3GA), the first 3D controllable model for human bodies rendered with Gaussian splats. Current photorealistic drivable avatars require either accurate 3D registrations during training, dense input images during testing, or both. The ones based on neural radiance fields also tend to be prohibitively slow for telepresence applications. This work uses the recently presented 3D Gaussian Splatting (3DGS) technique to render realistic humans at real-time framerates, using dense calibrated multi-view videos as input. To deform those primitives, we depart from the commonly used point deformation method of linear blend skinning (LBS) and use a classic volumetric deformation method: cage deformations. Given their smaller size, we drive these deformations with joint angles and keypoints, which are more suitable for communication applications. Our experiments on nine subjects with varied body shapes, clothes, and motions obtain higher-quality results than state-of-the-art methods when using the same training and test data.

AK

327,105 次观看 • 2 年前

We are thrilled to see that our recent Nature Biotechnology paper just became the most accessed publication in the journal within only three months (>225K accesses 🤩)—a fantastic achievement reflecting the power of teamwork! This paper has 7 first authors: two computer scientists, two biologists, and three chemists working together. It perfectly illustrates something I’ve learned deeply over the past 10 years leading my group: great science is mostly the synergy of diverse minds and skills. Building a successful team is like assembling Lego bricks—each person complements the others, fits precisely, and collectively forms something much greater than any single piece. Over the years, we’ve made mistakes and learned valuable lessons, particularly in finding team members whose values and work styles align with our team culture. Now, we’re more intentional about ensuring a mutual fit, making the collaboration enjoyable, reciprocal, and highly productive. This paper is a perfect example: by combining high-resolution imaging and AI, our interdisciplinary team enabled unprecedented visualization and therapeutic development at the single-cell level throughout entire organisms. Proud of the team effort behind this impactful research and deeply grateful to each member who made it possible! What are your experiences working in a highly collaborative team vs. better focusing on your own project? The latter is totally fine and works perfectly for many academic labs.

We are thrilled to see that our recent Nature Biotechnology paper just became the most accessed publication in the journal within only three months (>225K accesses 🤩)—a fantastic achievement reflecting the power of teamwork! This paper has 7 first authors: two computer scientists, two biologists, and three chemists working together. It perfectly illustrates something I’ve learned deeply over the past 10 years leading my group: great science is mostly the synergy of diverse minds and skills. Building a successful team is like assembling Lego bricks—each person complements the others, fits precisely, and collectively forms something much greater than any single piece. Over the years, we’ve made mistakes and learned valuable lessons, particularly in finding team members whose values and work styles align with our team culture. Now, we’re more intentional about ensuring a mutual fit, making the collaboration enjoyable, reciprocal, and highly productive. This paper is a perfect example: by combining high-resolution imaging and AI, our interdisciplinary team enabled unprecedented visualization and therapeutic development at the single-cell level throughout entire organisms. Proud of the team effort behind this impactful research and deeply grateful to each member who made it possible! What are your experiences working in a highly collaborative team vs. better focusing on your own project? The latter is totally fine and works perfectly for many academic labs.

Ali Max Erturk

19,678 次观看 • 1 年前

We knew from day one that unlocking constellation-scale manufacturing would be more than just a hardware challenge. That's why we built Octopus — our proprietary operating system, developed in-house to handle the many complexities of rapidly scaling satellite manufacturing. Octopus provides an end-to-end source of truth, from first customer conversation to orbit, enabling streamlined traceability that ensures speed never comes at the expense of quality. Octopus is built on Palantir's Foundry and AIP platforms, which provide the underlying infrastructure to rapidly prototype and deploy custom applications across the entire company. By building Octopus on Palantir, we have unified scheduling, procurement, receiving, inspection, inventory management, and production into a single integrated workflow — enabling predictable and secure, constellation-scale production. Hear from Apex CEO, Ian Cinnamon, and the rest of the team on how Apex's partnership with Palantir is unlocking rate production at Factory One, and laying the foundation for the next generation of satellite manufacturing.

We knew from day one that unlocking constellation-scale manufacturing would be more than just a hardware challenge. That's why we built Octopus — our proprietary operating system, developed in-house to handle the many complexities of rapidly scaling satellite manufacturing. Octopus provides an end-to-end source of truth, from first customer conversation to orbit, enabling streamlined traceability that ensures speed never comes at the expense of quality. Octopus is built on Palantir's Foundry and AIP platforms, which provide the underlying infrastructure to rapidly prototype and deploy custom applications across the entire company. By building Octopus on Palantir, we have unified scheduling, procurement, receiving, inspection, inventory management, and production into a single integrated workflow — enabling predictable and secure, constellation-scale production. Hear from Apex CEO, Ian Cinnamon, and the rest of the team on how Apex's partnership with Palantir is unlocking rate production at Factory One, and laying the foundation for the next generation of satellite manufacturing.

Apex - Satellite Platforms 🛰

33,843 次观看 • 5 个月前