Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

AutoAdapt automates the design and tuning of domain adaptation workflows for large language models. It improves performance without requiring additional compute, making deployment more accessible:

MSFTResearch's profile picture

Microsoft Research

554,476 subscribers

10,753 görüntüleme • 3 ay önce •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

Sam Altman "Internally, we've reached GPT-4.5, and getting to GPT-5.5 won't require 100x more compute" - progress in reasoning models and RL techniques have greatly improved compute efficiency - it allows smaller models like internal GPT-4.5 to achieve GPT-6 level performance without requiring 100x more computing power
1:09

Sam Altman "Internally, we've reached GPT-4.5, and getting to GPT-5.5 won't require 100x more compute" - progress in reasoning models and RL techniques have greatly improved compute efficiency - it allows smaller models like internal GPT-4.5 to achieve GPT-6 level performance without requiring 100x more computing power

Haider.

344,581 görüntüleme • 1 yıl önce

Introducing Computer for Enterprise Computer runs multi-step workflows across research, coding, design, and deployment. It routes tasks across 20 specialized models and connects to 400+ applications.
1:12

Introducing Computer for Enterprise Computer runs multi-step workflows across research, coding, design, and deployment. It routes tasks across 20 specialized models and connects to 400+ applications.

Perplexity

202,694 görüntüleme • 4 ay önce

we're turning University of Waterloo into a supercomputer! arceus is a cross-device distributed compute network for training large models, using model/tensor/pipeline parallelism. you can train anything, from deep neural networks to language models on the network. oss & deployment soon!
0:33

we're turning University of Waterloo into a supercomputer! arceus is a cross-device distributed compute network for training large models, using model/tensor/pipeline parallelism. you can train anything, from deep neural networks to language models on the network. oss & deployment soon!

rajan agarwal

35,720 görüntüleme • 1 yıl önce

AI is changing design and 3D printing, making it easier for anyone to create and sell products worldwide. The future of making is here—and it's more accessible than ever. #AI #3DPrinting
0:22

AI is changing design and 3D printing, making it easier for anyone to create and sell products worldwide. The future of making is here—and it's more accessible than ever. #AI #3DPrinting

3DOS | Decentralized Manufacturing Network

45,733 görüntüleme • 8 ay önce

Fine-tune FunctionGemma to resolve tool selection ambiguity and build more accurate agentic workflows. The FunctionGemma Tuning Lab lets you train on your own dataset and test natural language to software actions without writing code or managing infrastructure.
0:59

Fine-tune FunctionGemma to resolve tool selection ambiguity and build more accurate agentic workflows. The FunctionGemma Tuning Lab lets you train on your own dataset and test natural language to software actions without writing code or managing infrastructure.

Google AI Developers

20,634 görüntüleme • 6 ay önce

Introducing Parallax, the first fully distributed inference and serving engine for large language models. Try it now: 🧵
0:26

Introducing Parallax, the first fully distributed inference and serving engine for large language models. Try it now: 🧵

Gradient

161,112 görüntüleme • 1 yıl önce

TIPS: Besides 3D wallpaper, the latest XR Animator also improves the performance of some Transformers-based AI models. For example, object detection AI based on Transformers is slower but more accurate, and this update makes these models more usable💨
0:29

TIPS: Besides 3D wallpaper, the latest XR Animator also improves the performance of some Transformers-based AI models. For example, object detection AI based on Transformers is slower but more accurate, and this update makes these models more usable💨

Butz Yung

23,932 görüntüleme • 8 ay önce

The introduction of Workflows, Model Fine-tuning, Apps for Advertising and more. Get caught up on what happened This Week with Runway.
2:58

The introduction of Workflows, Model Fine-tuning, Apps for Advertising and more. Get caught up on what happened This Week with Runway.

Runway

18,527 görüntüleme • 9 ay önce

What we’re seeing: frontier robotics teams are rebuilding data pipelines from scratch because generic licensed datasets aren’t moving their models forward. The next wave won’t be about more data. It’ll be about precise, domain-specific data that actually improves performance. The bottleneck isn’t compute. It’s the right data. 🎥 Our CCO Arthur at Wrapped’s Multichain Day (ETHDenver 🏔🦬🦄)
0:43

What we’re seeing: frontier robotics teams are rebuilding data pipelines from scratch because generic licensed datasets aren’t moving their models forward. The next wave won’t be about more data. It’ll be about precise, domain-specific data that actually improves performance. The bottleneck isn’t compute. It’s the right data. 🎥 Our CCO Arthur at Wrapped’s Multichain Day (ETHDenver 🏔🦬🦄)

Perle Labs

30,597 görüntüleme • 5 ay önce

We got our robots to wash pans, clean windows, make peanut butter sandwiches, and more! Fine-tuning our latest model enables all of these tasks, and this has interesting implications for robotics, Moravec's paradox, and the future of large models in embodied AI. More below!
1:12

We got our robots to wash pans, clean windows, make peanut butter sandwiches, and more! Fine-tuning our latest model enables all of these tasks, and this has interesting implications for robotics, Moravec's paradox, and the future of large models in embodied AI. More below!

Physical Intelligence

543,784 görüntüleme • 7 ay önce

The lack of understanding of how current AI/ large language models work makes us susceptible to thinking that AI is more powerful and all-knowing than it actually is. That’s what concerns me the most.
2:09

The lack of understanding of how current AI/ large language models work makes us susceptible to thinking that AI is more powerful and all-knowing than it actually is. That’s what concerns me the most.

Sinead Bovell

143,642 görüntüleme • 3 yıl önce

⚡ GLTF Optimizer Optimize your 3D models for the web in seconds. Reduce file size, improve performance, and keep your #threejs scenes smooth all without losing quality. Perfect for faster 3D workflows. 👉 Creator: Simon #threejs #WebGL #3D #Optimization #Workflow #Performance #3DDev
0:41

⚡ GLTF Optimizer Optimize your 3D models for the web in seconds. Reduce file size, improve performance, and keep your #threejs scenes smooth all without losing quality. Perfect for faster 3D workflows. 👉 Creator: Simon #threejs #WebGL #3D #Optimization #Workflow #Performance #3DDev

Three.js Resources

18,141 görüntüleme • 6 ay önce

New Experiment: SerendipityLM - Interactive evolutionary exploration of generative design spaces with large language models. Check out the extensive experiment report, incl. many generative art examples & link to open source repo: Find out more in this 🧵
0:24

New Experiment: SerendipityLM - Interactive evolutionary exploration of generative design spaces with large language models. Check out the extensive experiment report, incl. many generative art examples & link to open source repo: Find out more in this 🧵

samim

23,739 görüntüleme • 2 yıl önce

Murray Shanahan emphasizes that cognitive ability and consciousness are distinct. An AI system can be highly intelligent—capable of goal achievement—without being conscious. This distinction is crucial when evaluating large language models, which lack awareness despite impressive performance.
1:51

Murray Shanahan emphasizes that cognitive ability and consciousness are distinct. An AI system can be highly intelligent—capable of goal achievement—without being conscious. This distinction is crucial when evaluating large language models, which lack awareness despite impressive performance.

Wes Roth

14,959 görüntüleme • 1 yıl önce

📈 NVIDIA tops AI leaderboards and benchmarks with open models driven by extreme co-design across compute, networking, memory, storage, and software. This includes models for biology, AI physics, agentic AI, physical AI, robotics, and autonomous vehicles. By being vertically integrated and horizontally open, NVIDIA enables its ecosystem to scale agentic workflows at every layer of the stack with the lowest cost per token.
0:38

📈 NVIDIA tops AI leaderboards and benchmarks with open models driven by extreme co-design across compute, networking, memory, storage, and software. This includes models for biology, AI physics, agentic AI, physical AI, robotics, and autonomous vehicles. By being vertically integrated and horizontally open, NVIDIA enables its ecosystem to scale agentic workflows at every layer of the stack with the lowest cost per token.

NVIDIA

85,541 görüntüleme • 3 ay önce

To Adapt or Not to Adapt? Real-Time Adaptation for Semantic Segmentation paper page: The goal of Online Domain Adaptation for semantic segmentation is to handle unforeseeable domain changes that occur during deployment, like sudden weather events. However, the high computational costs associated with brute-force adaptation make this paradigm unfeasible for real-world applications. In this paper we propose HAMLET, a Hardware-Aware Modular Least Expensive Training framework for real-time domain adaptation. Our approach includes a hardware-aware back-propagation orchestration agent (HAMT) and a dedicated domain-shift detector that enables active control over when and how the model is adapted (LT). Thanks to these advancements, our approach is capable of performing semantic segmentation while simultaneously adapting at more than 29FPS on a single consumer-grade GPU. Our framework's encouraging accuracy and speed trade-off is demonstrated on OnDA and SHIFT benchmarks through experimental results.
4:47

To Adapt or Not to Adapt? Real-Time Adaptation for Semantic Segmentation paper page: The goal of Online Domain Adaptation for semantic segmentation is to handle unforeseeable domain changes that occur during deployment, like sudden weather events. However, the high computational costs associated with brute-force adaptation make this paradigm unfeasible for real-world applications. In this paper we propose HAMLET, a Hardware-Aware Modular Least Expensive Training framework for real-time domain adaptation. Our approach includes a hardware-aware back-propagation orchestration agent (HAMT) and a dedicated domain-shift detector that enables active control over when and how the model is adapted (LT). Thanks to these advancements, our approach is capable of performing semantic segmentation while simultaneously adapting at more than 29FPS on a single consumer-grade GPU. Our framework's encouraging accuracy and speed trade-off is demonstrated on OnDA and SHIFT benchmarks through experimental results.

AK

18,871 görüntüleme • 3 yıl önce

Amera automates claims processing workflows for health insurance payers. It cuts admin costs by 20-30%, turns hours of manual work into minutes, and helps members access more affordable, transparent care. Congrats on the launch, Deep Kapur & Louise Tanski!
1:37

Amera automates claims processing workflows for health insurance payers. It cuts admin costs by 20-30%, turns hours of manual work into minutes, and helps members access more affordable, transparent care. Congrats on the launch, Deep Kapur & Louise Tanski!

Y Combinator

23,800 görüntüleme • 9 ay önce

Demis Hassabis says despite the advent of models like DeepSeek, AI needs "more compute than ever" for experimentation and innovation, serving models and inference because "the more computing time you spend, you get more powerful and better answers"
0:48

Demis Hassabis says despite the advent of models like DeepSeek, AI needs "more compute than ever" for experimentation and innovation, serving models and inference because "the more computing time you spend, you get more powerful and better answers"

Tsarathustra

53,451 görüntüleme • 1 yıl önce

🚀 Excited to announce the first release of a novel open source programming language and platform for language model interaction! Combining prompts, constraints & scripting, LMQL elevates the capabilities of large language models. 🧵1/6 A quick tour.
0:20

🚀 Excited to announce the first release of a novel open source programming language and platform for language model interaction! Combining prompts, constraints & scripting, LMQL elevates the capabilities of large language models. 🧵1/6 A quick tour.

LMQL (Language Model Query Language)

198,966 görüntüleme • 3 yıl önce

Google presents AudioPaLM: A Large Language Model That Can Speak and Listen paper page: introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models, PaLM-2 [Anil et al., 2023] and AudioLM [Borsos et al., 2022], into a unified multimodal architecture that can process and generate text and speech with applications including speech recognition and speech-to-speech translation. AudioPaLM inherits the capability to preserve paralinguistic information such as speaker identity and intonation from AudioLM and the linguistic knowledge present only in text large language models such as PaLM-2. We demonstrate that initializing AudioPaLM with the weights of a text-only large language model improves speech processing, successfully leveraging the larger quantity of text training data used in pretraining to assist with the speech tasks. The resulting model significantly outperforms existing systems for speech translation tasks and has the ability to perform zero-shot speech-to-text translation for many languages for which input/target language combinations were not seen in training. AudioPaLM also demonstrates features of audio language models, such as transferring a voice across languages based on a short spoken prompt.
4:10

Google presents AudioPaLM: A Large Language Model That Can Speak and Listen paper page: introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models, PaLM-2 [Anil et al., 2023] and AudioLM [Borsos et al., 2022], into a unified multimodal architecture that can process and generate text and speech with applications including speech recognition and speech-to-speech translation. AudioPaLM inherits the capability to preserve paralinguistic information such as speaker identity and intonation from AudioLM and the linguistic knowledge present only in text large language models such as PaLM-2. We demonstrate that initializing AudioPaLM with the weights of a text-only large language model improves speech processing, successfully leveraging the larger quantity of text training data used in pretraining to assist with the speech tasks. The resulting model significantly outperforms existing systems for speech translation tasks and has the ability to perform zero-shot speech-to-text translation for many languages for which input/target language combinations were not seen in training. AudioPaLM also demonstrates features of audio language models, such as transferring a voice across languages based on a short spoken prompt.

AK

290,517 görüntüleme • 3 yıl önce