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。

Reasoning = step-by-step thinking. Usually. Diffusion reasons in parallel and refines natively. Better quality without the latency tax. Samar Khanna on how diffusion reasoning works. We're hiring →

Inception

19,163 subscribers

16,490 次观看 • 6 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

3 条评论

Aida 的头像

Aida6 个月前

@samar_a_khanna Isn't diffusion more FLOPS heavy than AR? How come it works faster than AR then. I understand the hardware utilization is going to be higher, but with 20x faster generation, you need at least 20x higher hardware utilization to get the speed up.

HTV 的头像

HTV6 个月前

@samar_a_khanna Idea and speed is super cool,

Ranko Mosic 的头像

Ranko Mosic6 个月前

@samar_a_khanna

相关视频

Listen to Samar Khanna explain why parallel generation, rather than sequential, raises the performance ceiling for language models. Learn more about diffusion LLMs. → We're hiring:

Listen to Samar Khanna explain why parallel generation, rather than sequential, raises the performance ceiling for language models. Learn more about diffusion LLMs. → We're hiring:

Inception

18,582 次观看 • 7 个月前

We're moving beyond autoregressive LLMs! Autoregressive LLMs generate text word-by-word, which can be slow and affect quality, while diffusion models refine noise step-by-step, allowing for faster iterations and error correction. Here's Gemini Diffusion running at 857 tokens/s:

We're moving beyond autoregressive LLMs! Autoregressive LLMs generate text word-by-word, which can be slow and affect quality, while diffusion models refine noise step-by-step, allowing for faster iterations and error correction. Here's Gemini Diffusion running at 857 tokens/s:

Akshay 🚀

34,524 次观看 • 1 年前

Stable Diffusion Moment of Audio?? Ace-Step Audio Model just dropped and is now natively supported in ComfyUI. ACE-Step is an open-source music generation model jointly developed by ACE Studio and StepFun.

Stable Diffusion Moment of Audio?? Ace-Step Audio Model just dropped and is now natively supported in ComfyUI. ACE-Step is an open-source music generation model jointly developed by ACE Studio and StepFun.

ComfyUI

23,611 次观看 • 1 年前

NVIDIA just released AnyFlow on Hugging Face The first any-step video diffusion model that generates high-quality text-to-video with any inference budget - 4 steps or 50, quality scales smoothly without degradation.

NVIDIA just released AnyFlow on Hugging Face The first any-step video diffusion model that generates high-quality text-to-video with any inference budget - 4 steps or 50, quality scales smoothly without degradation.

DailyPapers

41,777 次观看 • 4 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

83,681 次观看 • 3 年前

🚨xAI: GROK 3 ALLOWS YOU TO FOLLOW ITS REASONING STEP-BY-STEP "Part of Grok's advanced reasoning capabilities are these thinking traces that you can see. You can even go inside and actually read what Grok is thinking as it's going through the problem, as it's trying to solve it." Source: xAI Elon Musk

🚨xAI: GROK 3 ALLOWS YOU TO FOLLOW ITS REASONING STEP-BY-STEP "Part of Grok's advanced reasoning capabilities are these thinking traces that you can see. You can even go inside and actually read what Grok is thinking as it's going through the problem, as it's trying to solve it." Source: xAI Elon Musk

Mario Nawfal

88,044 次观看 • 1 年前

Video might be the next intelligence substrate. Strikingly, video models are beginning to exhibit the same emergent reasoning behaviors first observed in LLMs—multi-path search, self-correction, and layer specialization. We demystify video reasoning and show it doesn’t happen frame-by-frame, but along diffusion steps. 🔗 📄 So, what's next? ;)

Video might be the next intelligence substrate. Strikingly, video models are beginning to exhibit the same emergent reasoning behaviors first observed in LLMs—multi-path search, self-correction, and layer specialization. We demystify video reasoning and show it doesn’t happen frame-by-frame, but along diffusion steps. 🔗 📄 So, what's next? ;)

Zhongang Cai

71,080 次观看 • 6 个月前

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

18,196 次观看 • 11 个月前

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

30,446 次观看 • 2 年前

Interactive Reasoning Benchmarks are the next step in frontier evaluations Hear Greg Kamradt share why measuring human-like intelligence requires multi-turn environments Including a sneak peak of ARC-AGI-3 Want to help us build interactive evaluations? We're hiring

Interactive Reasoning Benchmarks are the next step in frontier evaluations Hear Greg Kamradt share why measuring human-like intelligence requires multi-turn environments Including a sneak peak of ARC-AGI-3 Want to help us build interactive evaluations? We're hiring

ARC Prize

27,427 次观看 • 1 年前

A 10B Chinese model beats models 20x its size. And it runs on your laptop. Step 3 VL10B uses parallel reasoning to think like 16 experts at once. → Beats 200B parameter models → Scores 92.2% on multimodal benchmarks → Runs 16 reasoning paths simultaneously → Open source on Hugging Face → Works offline on your hardware OCR. GUI understanding. STEM problems. All covered. Completely free to download. Want the technical breakdown? DM me.

A 10B Chinese model beats models 20x its size. And it runs on your laptop. Step 3 VL10B uses parallel reasoning to think like 16 experts at once. → Beats 200B parameter models → Scores 92.2% on multimodal benchmarks → Runs 16 reasoning paths simultaneously → Open source on Hugging Face → Works offline on your hardware OCR. GUI understanding. STEM problems. All covered. Completely free to download. Want the technical breakdown? DM me.

Julian Goldie SEO

37,243 次观看 • 8 个月前

Code Arena can handle image inputs for agentic web dev tasks, reasoning through multi-step problems and using tools along the way. Watch how it works with Aryan Vichare in this clip. Find a link to the full walkthrough on how Code Arena creates sites and apps from images in thread.

Code Arena can handle image inputs for agentic web dev tasks, reasoning through multi-step problems and using tools along the way. Watch how it works with Aryan Vichare in this clip. Find a link to the full walkthrough on how Code Arena creates sites and apps from images in thread.

Arena.ai

20,064 次观看 • 6 个月前

Three new open-source models just landed in ComfyUI natively: → Gemma 4 (Google DeepMind) - multimodal LLM handling text, image, audio, and video input with built-in step-by-step reasoning mode → VOID (Netflix) - video object removal that also erases shadows, reflections, and physical interactions caused by the removed subject → BiRefNet - high-res background & object segmentation, one of the most-used segmentation models in the ecosystem Workflows and blog linked below 👇

Three new open-source models just landed in ComfyUI natively: → Gemma 4 (Google DeepMind) - multimodal LLM handling text, image, audio, and video input with built-in step-by-step reasoning mode → VOID (Netflix) - video object removal that also erases shadows, reflections, and physical interactions caused by the removed subject → BiRefNet - high-res background & object segmentation, one of the most-used segmentation models in the ecosystem Workflows and blog linked below 👇

ComfyUI

218,749 次观看 • 4 个月前

How can we get VLMs to move their eyes—and reason step-by-step in visually grounded ways? 👀 We introduce ViGoRL, a RL method that anchors reasoning to image regions. 🎯 It outperforms vanilla GRPO and SFT across grounding, spatial tasks, and visual search (86.4% on V*). 👇🧵

How can we get VLMs to move their eyes—and reason step-by-step in visually grounded ways? 👀 We introduce ViGoRL, a RL method that anchors reasoning to image regions. 🎯 It outperforms vanilla GRPO and SFT across grounding, spatial tasks, and visual search (86.4% on V*). 👇🧵

Gabriel Sarch

76,717 次观看 • 1 年前

I'm thrilled to announce the launch of ⚡️Flash Diffusion from Jasper! Earlier this year, with our acquisition of Clipdrop, we launched the Jasper AI Research Lab in Paris. Today, we are excited to release our first piece of groundbreaking research: the open-source distillation method, "Flash Diffusion". Flash Diffusion accelerates inference by 500%, reduces computing costs, and produces higher-quality image outputs. Dive into the details and discover how Flash Diffusion is set to revolutionize the field of AI and image synthesis. Read all about it here: Try a demo on Hugging Face:

I'm thrilled to announce the launch of ⚡️Flash Diffusion from Jasper! Earlier this year, with our acquisition of Clipdrop, we launched the Jasper AI Research Lab in Paris. Today, we are excited to release our first piece of groundbreaking research: the open-source distillation method, "Flash Diffusion". Flash Diffusion accelerates inference by 500%, reduces computing costs, and produces higher-quality image outputs. Dive into the details and discover how Flash Diffusion is set to revolutionize the field of AI and image synthesis. Read all about it here: Try a demo on Hugging Face:

Timothy Young

10,093 次观看 • 2 年前

Introducing typesafe/jev-router: a cache-aware model router powered by Jev and TypeSafe AI The Jev Router picks the best model and reasoning effort for each request, balancing quality, speed, and cost. Here's how it works 👇🏻

Introducing typesafe/jev-router: a cache-aware model router powered by Jev and TypeSafe AI The Jev Router picks the best model and reasoning effort for each request, balancing quality, speed, and cost. Here's how it works 👇🏻

OpenRouter

1,327,773 次观看 • 11 天前

How to halve your car & home insurance cost just by getting it on the right day (and why it works). Watch back the full @itvmlshow: - for the full step-by-step on how to beat the huge rise in insurance costs

How to halve your car & home insurance cost just by getting it on the right day (and why it works). Watch back the full @itvmlshow: - for the full step-by-step on how to beat the huge rise in insurance costs

Martin Lewis

1,030,375 次观看 • 2 年前

SGLang-Diffusion with VDN-H3 now generates 14.4s of 768p video in just 9.0s 🚀 On 8× B200, 8 step denoising takes just 6.9s, reaching over 2× real time. The 9.0s figure covers the full generation request after warmup. No measured quality regression versus dense 50-step H3 across 103 test prompts. 🧵

SGLang-Diffusion with VDN-H3 now generates 14.4s of 768p video in just 9.0s 🚀 On 8× B200, 8 step denoising takes just 6.9s, reaching over 2× real time. The 9.0s figure covers the full generation request after warmup. No measured quality regression versus dense 50-step H3 across 103 test prompts. 🧵

SGLang

80,104 次观看 • 22 天前

🚀 Introducing Emu3.5 — a large-scale multimodal world model that natively predicts the next vision-language state. 🔥 Trained on over 10T interleaved vision-language tokens and enhanced with reinforcement learning, Emu3.5 achieves powerful multimodal reasoning and generation. ⚡ Powered by our new Discrete Diffusion Adaptation (DiDA) for 20× faster inference. 🔥 Emu3.5 outperforms Nano Banana across image generation, editing, interleaved tasks and more. 🌍 Explore Emu3.5: Github: #Emu3 #MultimodalAI #WorldModel #NextTokenPrediction

🚀 Introducing Emu3.5 — a large-scale multimodal world model that natively predicts the next vision-language state. 🔥 Trained on over 10T interleaved vision-language tokens and enhanced with reinforcement learning, Emu3.5 achieves powerful multimodal reasoning and generation. ⚡ Powered by our new Discrete Diffusion Adaptation (DiDA) for 20× faster inference. 🔥 Emu3.5 outperforms Nano Banana across image generation, editing, interleaved tasks and more. 🌍 Explore Emu3.5: Github: #Emu3 #MultimodalAI #WorldModel #NextTokenPrediction

BAAI

51,880 次观看 • 11 个月前