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。

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 @CVPR'26

1,321 subscribers

70,629 次观看 • 4 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

Demis Hassabis on the next 12 Months: - Full multimodal convergence: Models like Gemini will seamlessly take in and output text, images, audio, and video, with cross-pollination that boosts reasoning + creativity. - Breakthrough visual intelligence: Image models like Nano Banana Pro will produce highly accurate infographics and show near-human visual understanding. - Language + video fusion: Video models integrated with LLMs unlock richer analysis, storytelling, and step-by-step visual reasoning. - World models go mainstream like Genie 3 - Agents become reliable

Demis Hassabis on the next 12 Months: - Full multimodal convergence: Models like Gemini will seamlessly take in and output text, images, audio, and video, with cross-pollination that boosts reasoning + creativity. - Breakthrough visual intelligence: Image models like Nano Banana Pro will produce highly accurate infographics and show near-human visual understanding. - Language + video fusion: Video models integrated with LLMs unlock richer analysis, storytelling, and step-by-step visual reasoning. - World models go mainstream like Genie 3 - Agents become reliable

Chubby♨️

89,236 次观看 • 7 个月前

AI has transformed how video is created. We think the next wave is about understanding it. Over the past few years, we've seen remarkable advances in video generation, editing, avatars, and creative tooling. An increasingly important problem is teaching machines to search, analyze, reason over, and extract insight from video - across massive libraries and live streams alike. We're calling this video intelligence, and we're actively looking to back founders building here. We're most excited about companies pushing on the core capabilities: - Video-native models - multimodal embeddings, temporal reasoning, and retrieval built specifically for video rather than adapted from image or text - Real-time and large-scale pipelines - infrastructure for processing, indexing, and querying video at the speed and scale enterprises actually need - Agentic and reasoning layers - systems that don't just retrieve clips but answer questions, surface anomalies, and take action on what they see The models and infrastructure to make this real are appearing to be crossing a capability threshold right now. Multimodal foundation models are maturing, storage costs have collapsed, and enterprises are sitting on years of unstructured video with no way to use it. That infrastructure unlocks a wide range of applications including media and sports workflows, security and physical operations, enterprise knowledge management, advertising analytics, robotics, and consumer products, where video has historically been dark data. If you're building in video intelligence at the model layer, the platform layer, or in a vertical application, we'd love to talk!

AI has transformed how video is created. We think the next wave is about understanding it. Over the past few years, we've seen remarkable advances in video generation, editing, avatars, and creative tooling. An increasingly important problem is teaching machines to search, analyze, reason over, and extract insight from video - across massive libraries and live streams alike. We're calling this video intelligence, and we're actively looking to back founders building here. We're most excited about companies pushing on the core capabilities: - Video-native models - multimodal embeddings, temporal reasoning, and retrieval built specifically for video rather than adapted from image or text - Real-time and large-scale pipelines - infrastructure for processing, indexing, and querying video at the speed and scale enterprises actually need - Agentic and reasoning layers - systems that don't just retrieve clips but answer questions, surface anomalies, and take action on what they see The models and infrastructure to make this real are appearing to be crossing a capability threshold right now. Multimodal foundation models are maturing, storage costs have collapsed, and enterprises are sitting on years of unstructured video with no way to use it. That infrastructure unlocks a wide range of applications including media and sports workflows, security and physical operations, enterprise knowledge management, advertising analytics, robotics, and consumer products, where video has historically been dark data. If you're building in video intelligence at the model layer, the platform layer, or in a vertical application, we'd love to talk!

Jason Cui

36,205 次观看 • 2 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

Goodnight, 𝕏..·˚ ༘ ☾ ⋆｡˚ ☄︎ Newscast about xAI, SpaceX, Starlink, Neuralink, 𝕏, and Grok. Created using the xAI Video API. I upscaled the final frame to use as the starting point for the next clip's generation, but it only helped marginally. However, I found a paper called "FreeInit: Bridging Initialization Gap in Video Diffusion Models" that I'm going to try tomorrow to fix the drifting problem. The approach: 1. FFT both images (anchor frame + last frame) 2. Low-pass filter on anchor (structure/color/layout) 3. High-pass filter on last frame (details/edges/motion) 4. Combine and IFFT back

Goodnight, 𝕏..·˚ ༘ ☾ ⋆｡˚ ☄︎ Newscast about xAI, SpaceX, Starlink, Neuralink, 𝕏, and Grok. Created using the xAI Video API. I upscaled the final frame to use as the starting point for the next clip's generation, but it only helped marginally. However, I found a paper called "FreeInit: Bridging Initialization Gap in Video Diffusion Models" that I'm going to try tomorrow to fix the drifting problem. The approach: 1. FFT both images (anchor frame + last frame) 2. Low-pass filter on anchor (structure/color/layout) 3. High-pass filter on last frame (details/edges/motion) 4. Combine and IFFT back

tetsuo

97,901 次观看 • 5 个月前

Test-time scaling, reasoning, and generally search-like processes clearly drive significant gains in LLMs. Largely owed to the structure of language. One would think the same could apply to non-linguistic domains, like image generation, but that obviously depends on whether the structure of the domain's representation lends itself to search. 1D ordered tokens (e.g., image FlexTok, video FlexTok) seem like a natural fit since they enable a step-by-step coarse-to-fine generation. We investigated that and found they indeed enable search and scale far better with test-time compute than 2D grids. See the visuals on the webpage. Appearing in ICML Conference 2026. 🔗 📄

Test-time scaling, reasoning, and generally search-like processes clearly drive significant gains in LLMs. Largely owed to the structure of language. One would think the same could apply to non-linguistic domains, like image generation, but that obviously depends on whether the structure of the domain's representation lends itself to search. 1D ordered tokens (e.g., image FlexTok, video FlexTok) seem like a natural fit since they enable a step-by-step coarse-to-fine generation. We investigated that and found they indeed enable search and scale far better with test-time compute than 2D grids. See the visuals on the webpage. Appearing in ICML Conference 2026. 🔗 📄

Amir Zamir

15,407 次观看 • 2 个月前

Our research team just released Flex-Forcing: a video generation method that lets a single model switch between generation methods at inference time. Right now there are two main approaches to video generation. Bidirectional diffusion models attend to every frame at once, holding structure well at the cost of speed. Autoregressive models generate frame by frame, so they stream fast and scale to long clips, but accumulate error and drift over time. Flex-Forcing trains a single model to do both, letting you choose from the range at inference based on your compute budget.

Our research team just released Flex-Forcing: a video generation method that lets a single model switch between generation methods at inference time. Right now there are two main approaches to video generation. Bidirectional diffusion models attend to every frame at once, holding structure well at the cost of speed. Autoregressive models generate frame by frame, so they stream fast and scale to long clips, but accumulate error and drift over time. Flex-Forcing trains a single model to do both, letting you choose from the range at inference based on your compute budget.

NVIDIA AI

31,412 次观看 • 7 天前

Some say UN is following a similar path to UNI and CAKE… but similar in what aspect? 🤔 We think “same same, but different” might be the best way to put it. Watch the video to see why 👇🏻

Some say UN is following a similar path to UNI and CAKE… but similar in what aspect? 🤔 We think “same same, but different” might be the best way to put it. Watch the video to see why 👇🏻

Unich.com 🆓

171,928 次观看 • 9 个月前

Computer vision can see what happened — agentic AI explains why it matters and what to do next. Here’s how teams are upgrading video analytics with vision language models: 🔍 Turn video into searchable intelligence 🧠 Add context and reasoning to system alerts ⚡ Summarize complex scenes and answer questions automatically Read the blog to see 3 real-world ways to bring agentic AI to computer vision →

Computer vision can see what happened — agentic AI explains why it matters and what to do next. Here’s how teams are upgrading video analytics with vision language models: 🔍 Turn video into searchable intelligence 🧠 Add context and reasoning to system alerts ⚡ Summarize complex scenes and answer questions automatically Read the blog to see 3 real-world ways to bring agentic AI to computer vision →

NVIDIA AI

14,273 次观看 • 6 个月前

Hiring RL Engineer! Started off as a curious project at Lossfunk to push the boundaries of LLMs in social reasoning - we are now building RL environments, data, and benchmarks to simulate more real-world scenarios. If you want to train SoTA RL models over multi-GPUs (H200s/B200s) to unlock next AI frontier, this is for you.

Hiring RL Engineer! Started off as a curious project at Lossfunk to push the boundaries of LLMs in social reasoning - we are now building RL environments, data, and benchmarks to simulate more real-world scenarios. If you want to train SoTA RL models over multi-GPUs (H200s/B200s) to unlock next AI frontier, this is for you.

Satpal Singh Rathore

45,974 次观看 • 11 个月前

V3D Video Diffusion Models are Effective 3D Generators Automatic 3D generation has recently attracted widespread attention. Recent methods have greatly accelerated the generation speed, but usually produce less-detailed objects due to limited model capacity or 3D data. Motivated by recent advancements in video diffusion models, we introduce V3D, which leverages the world simulation capacity of pre-trained video diffusion models to facilitate 3D generation. To fully unleash the potential of video diffusion to perceive the 3D world, we further introduce geometrical consistency prior and extend the video diffusion model to a multi-view consistent 3D generator. Benefiting from this, the state-of-the-art video diffusion model could be fine-tuned to generate 360degree orbit frames surrounding an object given a single image. With our tailored reconstruction pipelines, we can generate high-quality meshes or 3D Gaussians within 3 minutes. Furthermore, our method can be extended to scene-level novel view synthesis, achieving precise control over the camera path with sparse input views. Extensive experiments demonstrate the superior performance of the proposed approach, especially in terms of generation quality and multi-view consistency

V3D Video Diffusion Models are Effective 3D Generators Automatic 3D generation has recently attracted widespread attention. Recent methods have greatly accelerated the generation speed, but usually produce less-detailed objects due to limited model capacity or 3D data. Motivated by recent advancements in video diffusion models, we introduce V3D, which leverages the world simulation capacity of pre-trained video diffusion models to facilitate 3D generation. To fully unleash the potential of video diffusion to perceive the 3D world, we further introduce geometrical consistency prior and extend the video diffusion model to a multi-view consistent 3D generator. Benefiting from this, the state-of-the-art video diffusion model could be fine-tuned to generate 360degree orbit frames surrounding an object given a single image. With our tailored reconstruction pipelines, we can generate high-quality meshes or 3D Gaussians within 3 minutes. Furthermore, our method can be extended to scene-level novel view synthesis, achieving precise control over the camera path with sparse input views. Extensive experiments demonstrate the superior performance of the proposed approach, especially in terms of generation quality and multi-view consistency

AK

31,997 次观看 • 2 年前

Economic experts have been warning us about the next financial crisis since 2018. I gave up waiting for it to happen but this might be a sign that we are getting very close. Check out this video by Matt Baker.

Economic experts have been warning us about the next financial crisis since 2018. I gave up waiting for it to happen but this might be a sign that we are getting very close. Check out this video by Matt Baker.

RyanMatta 🇺🇸 🦅

250,160 次观看 • 2 个月前

🔎 Frame-by-frame breakdown of ProPixel Video Analysis and Research's video: Shows not ONE but TWO distinct interdimensional beings: possibly a short Grey and a tall Grey. Time-dilation effect makes the taller figure appear to move faster (Hal Puthoff predicted this). We see UFOs zip by, they see us in slow motion. The Tall Figure appears to be wearing a wrist device and a large belt. Common attire in abduction accounts. Psionic session by Michael Battista. This is next-level. Not one entity — TWO. 👽

🔎 Frame-by-frame breakdown of ProPixel Video Analysis and Research's video: Shows not ONE but TWO distinct interdimensional beings: possibly a short Grey and a tall Grey. Time-dilation effect makes the taller figure appear to move faster (Hal Puthoff predicted this). We see UFOs zip by, they see us in slow motion. The Tall Figure appears to be wearing a wrist device and a large belt. Common attire in abduction accounts. Psionic session by Michael Battista. This is next-level. Not one entity — TWO. 👽

High Strangeness

78,764 次观看 • 23 天前

The DeepSeek-R1 paper is a gem! Highly encourage everyone to read it. It's clear that LLM reasoning capabilities can be learned in different ways. RL, if applied correctly and at scale, can lead to some really powerful and interesting scaling and emergent properties. There is more to RL than meets the eye! Here is my breakdown of the paper along with a few tests: The multi-state training might not make sense initially but they provide clues on optimizations that we can continue to tap into. Data quality is still very important for enhancing the usability of the LLM. Unlike other reasoning LLMs, DeepSeek-R1's training recipe and weights are open so we can build on top of it. This opens up exciting research opportunities. About the attached clip: the previous preview model wasn't able to solve this task. DeepSeek-R1 can solve this and many other tasks that o1 can solve. It's a very good model for coding and math.

The DeepSeek-R1 paper is a gem! Highly encourage everyone to read it. It's clear that LLM reasoning capabilities can be learned in different ways. RL, if applied correctly and at scale, can lead to some really powerful and interesting scaling and emergent properties. There is more to RL than meets the eye! Here is my breakdown of the paper along with a few tests: The multi-state training might not make sense initially but they provide clues on optimizations that we can continue to tap into. Data quality is still very important for enhancing the usability of the LLM. Unlike other reasoning LLMs, DeepSeek-R1's training recipe and weights are open so we can build on top of it. This opens up exciting research opportunities. About the attached clip: the previous preview model wasn't able to solve this task. DeepSeek-R1 can solve this and many other tasks that o1 can solve. It's a very good model for coding and math.

elvis

140,692 次观看 • 1 年前

Introducing iMessage video ads in Claude We taught Claude to make iMessage video ads in one shot with a single skill. These iMessage ads are absolutely killing it on Meta right now. The skill teaches Claude to make a full video ad in one shot with iPhone frame, SFX, music, and end card. And it's just HTML. It doesn't use any video generation models, so it's cheap! Comment Goose below and I'll send you the skill.

Introducing iMessage video ads in Claude We taught Claude to make iMessage video ads in one shot with a single skill. These iMessage ads are absolutely killing it on Meta right now. The skill teaches Claude to make a full video ad in one shot with iPhone frame, SFX, music, and end card. And it's just HTML. It doesn't use any video generation models, so it's cheap! Comment Goose below and I'll send you the skill.

Shiv

235,877 次观看 • 9 天前

Introducing The Matrix --- a foundation world model for generating infinite-length, hyper-realistic videos with real-time, frame-level control: - Infinite-length video generation - 720p high-quality rendering - Real-time, frame-level control at 16 FPS - Generalization to real-world video control 🔗Blog: 📄Paper: 💻Code & Playable Demo: Coming soon! Key Innovation: A brand new technique called the shift-window denoise process model, enabling auto-regressive generation for diffusion and consistency models in real-time. Special thanks to project leader Ruili Feng and the entire Matrix team for their dedication and hard work over the year-long project.

Hongyang Zhang

178,322 次观看 • 1 年前

If you let them ban the AR-15, they'll be trying to ban handguns next. If you let them ban handguns, they'll be trying to ban knives next. People always say that I'm crazy and just fear-mongering, but it's simple deductive reasoning. More people are killed with handguns and knives than with AR-15s. So, if they're trying to ban AR-15s under the guise of saving lives, what's to stop them from using the same justification to ban handguns and knives next? The answer is nothing. But for those who lack such levels of reductive reasoning, allow me to direct your attention to what they're talking about in the UK right now. Banning so-called 'Zombie' knives. Watch to learn more. AR-15s are Protected by the 2nd Amendment Embroidered Premium Hoodie #ThePewPewLife

If you let them ban the AR-15, they'll be trying to ban handguns next. If you let them ban handguns, they'll be trying to ban knives next. People always say that I'm crazy and just fear-mongering, but it's simple deductive reasoning. More people are killed with handguns and knives than with AR-15s. So, if they're trying to ban AR-15s under the guise of saving lives, what's to stop them from using the same justification to ban handguns and knives next? The answer is nothing. But for those who lack such levels of reductive reasoning, allow me to direct your attention to what they're talking about in the UK right now. Banning so-called 'Zombie' knives. Watch to learn more. AR-15s are Protected by the 2nd Amendment Embroidered Premium Hoodie #ThePewPewLife

Colion Noir

174,691 次观看 • 2 年前

‼️FULL 11-Minute TPUSA People's Convention In Detroit Is Going VIRAL But NOT For Reason You Think—What's At The 7-minute Mark Changes EVERYTHING 👀 Everyone is talking about Nick Fuentes and Jake Shields being removed, but there is so much more happening in this frame than the "mainstream" alt-media is reporting. Watch the full, uncut video here. I’ve been scrubbing this frame-by-frame and what I found in the background changes everything. Full breakdown of the anomalies coming in the next post. STAY TUNED. 👇 FROM: Matthew St Baker FOLLOW HIM!

‼️FULL 11-Minute TPUSA People's Convention In Detroit Is Going VIRAL But NOT For Reason You Think—What's At The 7-minute Mark Changes EVERYTHING 👀 Everyone is talking about Nick Fuentes and Jake Shields being removed, but there is so much more happening in this frame than the "mainstream" alt-media is reporting. Watch the full, uncut video here. I’ve been scrubbing this frame-by-frame and what I found in the background changes everything. Full breakdown of the anomalies coming in the next post. STAY TUNED. 👇 FROM: Matthew St Baker FOLLOW HIM!

Project Constitution

111,017 次观看 • 2 个月前

How to set up your Luel account and start earning up to $20/hr once your recordings are accepted. The steps are simple—just follow everything as shown in the video. In my next video, I’ll show you how to connect your account to Stripe so you can get paid easily. Make sure you’re following me so you don’t miss it.

How to set up your Luel account and start earning up to $20/hr once your recordings are accepted. The steps are simple—just follow everything as shown in the video. In my next video, I’ll show you how to connect your account to Stripe so you can get paid easily. Make sure you’re following me so you don’t miss it.

Ayilola of Nigeria /Laptop Guy 🩺 🇳🇬 🇬🇧🇺🇸

17,970 次观看 • 2 个月前

🚀 New Paper: Pixel Reasoner 🧠🖼️ How can Vision-Language Models (VLMs) perform chain-of-thought reasoning within the image itself? We introduce Pixel Reasoner, the first open-source framework that enables VLMs to “think in pixel space” through curiosity-driven reinforcement learning. Current VLMs reason only in text — even when grounded in rich images or videos, their logical steps are verbalized in natural language. This restricts their ability to interrogate visual evidence and demonstrate how conclusions are drawn. 🔍 So we ask: What if we could make VLMs "show their work" by reasoning directly in the pixel space? Inspired by GPT-o3’s "think-in-image" ability, we propose a framework where VLMs use interactive visual operations — zoom, select-frame, highlight — to reason through complex visual inputs. To do this, we design a two-stage training process: Instruction tuning with synthesized visual reasoning traces. Reinforcement learning with curiosity-driven reward to balance exploration between pixel and text reasoning ✨ With this, Pixel Reasoner achieves near-SoTA performance on many information-rich multimodal benchmarks: 📊 84% on InfographicsVQA 🧠 84% on V* benchmark 🧩 74% on TallyQA-Complex It also achieves strong accuracy of 68% on MVBench (a video benchmark). Website: Paper: Code: Demo: (coming soon)

🚀 New Paper: Pixel Reasoner 🧠🖼️ How can Vision-Language Models (VLMs) perform chain-of-thought reasoning within the image itself? We introduce Pixel Reasoner, the first open-source framework that enables VLMs to “think in pixel space” through curiosity-driven reinforcement learning. Current VLMs reason only in text — even when grounded in rich images or videos, their logical steps are verbalized in natural language. This restricts their ability to interrogate visual evidence and demonstrate how conclusions are drawn. 🔍 So we ask: What if we could make VLMs "show their work" by reasoning directly in the pixel space? Inspired by GPT-o3’s "think-in-image" ability, we propose a framework where VLMs use interactive visual operations — zoom, select-frame, highlight — to reason through complex visual inputs. To do this, we design a two-stage training process: Instruction tuning with synthesized visual reasoning traces. Reinforcement learning with curiosity-driven reward to balance exploration between pixel and text reasoning ✨ With this, Pixel Reasoner achieves near-SoTA performance on many information-rich multimodal benchmarks: 📊 84% on InfographicsVQA 🧠 84% on V* benchmark 🧩 74% on TallyQA-Complex It also achieves strong accuracy of 68% on MVBench (a video benchmark). Website: Paper: Code: Demo: (coming soon)

Wenhu Chen

82,829 次观看 • 1 年前