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。

VFXMaster Unlocking Dynamic Visual Effect Generation via In-Context Learning

AK

492,763 subscribers

29,059 次观看 • 9 个月前 •via X (Twitter)

教育新闻政治科学技术

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

The visual effect was really in Black Adam

The visual effect was really in Black Adam

Unique Movie Moments 🐬

99,094 次观看 • 1 年前

This is a single uncut video, showing a robot learning several tasks instantly, after just one demonstration each ... This is possible because we've now been able to achieve in-context learning for everyday robotics tasks, and I'm very excited to announce our latest paper: 🎆 Instant Policy: In-Context Imitation Learning via Graph Diffusion 🎆 (1/6) 🧵👇

This is a single uncut video, showing a robot learning several tasks instantly, after just one demonstration each ... This is possible because we've now been able to achieve in-context learning for everyday robotics tasks, and I'm very excited to announce our latest paper: 🎆 Instant Policy: In-Context Imitation Learning via Graph Diffusion 🎆 (1/6) 🧵👇

Edward Johns

74,683 次观看 • 1 年前

Vision-language models perform diverse tasks via in-context learning. Time for robots to do the same! Introducing In-Context Robot Transformer (ICRT): a robot policy that learns new tasks by prompting with robot trajectories, without any fine-tuning. [1/N]

Vision-language models perform diverse tasks via in-context learning. Time for robots to do the same! Introducing In-Context Robot Transformer (ICRT): a robot policy that learns new tasks by prompting with robot trajectories, without any fine-tuning. [1/N]

Max Fu

40,466 次观看 • 1 年前

new dynamic duo in NY? 🤔 (via 7PM in Brooklyn)

new dynamic duo in NY? 🤔 (via 7PM in Brooklyn)

NFL

232,553 次观看 • 1 年前

Have you tried the integrated terminal suggestions feature in Visual Studio Code? It gives you lightning-fast completions, like dynamic branch suggestions or path navigation. Available via the "terminal.integrated.suggest.enabled" setting (enabled by default in Insiders)

Have you tried the integrated terminal suggestions feature in Visual Studio Code? It gives you lightning-fast completions, like dynamic branch suggestions or path navigation. Available via the "terminal.integrated.suggest.enabled" setting (enabled by default in Insiders)

Visual Studio Code

109,948 次观看 • 1 年前

Gemini Deep Research just went visual. Gemini’s latest update turns dense topics into vivid, intuitive learning moments. Now when you dive deep on a subject, you don’t just read, you see: 🔹Charts 🔹 Diagrams 🔹 Animations 🔹 Visual context that sticks

Gemini Deep Research just went visual. Gemini’s latest update turns dense topics into vivid, intuitive learning moments. Now when you dive deep on a subject, you don’t just read, you see: 🔹Charts 🔹 Diagrams 🔹 Animations 🔹 Visual context that sticks

Wes Roth

37,290 次观看 • 7 个月前

Hillary Clinton calls out the real issue: a generation learning about Israel and Oct. 7 from TikTok instead of history. When context disappears, propaganda fills the gap — and it shows in the slogans students can’t even explain.

Hillary Clinton calls out the real issue: a generation learning about Israel and Oct. 7 from TikTok instead of history. When context disappears, propaganda fills the gap — and it shows in the slogans students can’t even explain.

HonestReporting

26,894 次观看 • 8 个月前

"Flashed Face Distortion Effect," a visual illusion where rapidly presented faces in peripheral vision appear distorted

"Flashed Face Distortion Effect," a visual illusion where rapidly presented faces in peripheral vision appear distorted

Science girl

198,872 次观看 • 10 个月前

✨ Introducing Keypoint Action Tokens. 🤖 We translate visual observations and robot actions into a "language" that off-the-shelf LLMs can ingest and output. This transforms LLMs into *in-context, low-level imitation learning machines*. 🚀 Let me explain. 👇🧵

✨ Introducing Keypoint Action Tokens. 🤖 We translate visual observations and robot actions into a "language" that off-the-shelf LLMs can ingest and output. This transforms LLMs into in-context, low-level imitation learning machines. 🚀 Let me explain. 👇🧵

Norman Di Palo

23,097 次观看 • 2 年前

ByteDance just dropped UNO on Hugging Face Less-to-More Generalization Unlocking More Controllability by In-Context Generation a universal framework that evolves from single-subject to multi-subject customization. UNO demonstrates strong generalization capabilities and is capable of unifying diverse tasks under one model

ByteDance just dropped UNO on Hugging Face Less-to-More Generalization Unlocking More Controllability by In-Context Generation a universal framework that evolves from single-subject to multi-subject customization. UNO demonstrates strong generalization capabilities and is capable of unifying diverse tasks under one model

AK

82,709 次观看 • 1 年前

Gemini 3 brings new capabilities to Google Gemini including visual layout and dynamic view for more visual and interactive responses. Check it out in action with these 3 prompts:

Gemini 3 brings new capabilities to Google Gemini including visual layout and dynamic view for more visual and interactive responses. Check it out in action with these 3 prompts:

Google

75,222 次观看 • 8 个月前

FramePack is out Packing Input Frame Context in Next-Frame Prediction Models for Video Generation

FramePack is out Packing Input Frame Context in Next-Frame Prediction Models for Video Generation

AK

31,486 次观看 • 1 年前

Great visual of NVIDIA’s Inference Context Memory architecture. KV cache offload to NVMe via BlueField-4 DPU this is how Rubin handles massive context windows at scale. Similar to Amazon’s approach but with full-stack hardware integration. H/T 程工

Great visual of NVIDIA’s Inference Context Memory architecture. KV cache offload to NVMe via BlueField-4 DPU this is how Rubin handles massive context windows at scale. Similar to Amazon’s approach but with full-stack hardware integration. H/T 程工

Ben Pouladian

54,335 次观看 • 6 个月前

How to scale visual affordance learning that is fine-grained, task-conditioned, works in-the-wild, in dynamic envs? Introducing Unsupervised Affordance Distillation (UAD): distills affordances from off-the-shelf foundation models, *all without manual labels*. Very excited this is nominated as Best Paper Finalist at #ICRA2025! 🧵👇

How to scale visual affordance learning that is fine-grained, task-conditioned, works in-the-wild, in dynamic envs? Introducing Unsupervised Affordance Distillation (UAD): distills affordances from off-the-shelf foundation models, all without manual labels. Very excited this is nominated as Best Paper Finalist at #ICRA2025! 🧵👇

Wenlong Huang

93,662 次观看 • 1 年前

Starlink will connect students in The Bahamas to reliable high-speed internet, empowering online learning and unlocking a universe of potential 🛰️❤️ →

Starlink will connect students in The Bahamas to reliable high-speed internet, empowering online learning and unlocking a universe of potential 🛰️❤️ →

Starlink

689,545 次观看 • 7 个月前