Исследования1 мин
泰坦：在考试时学习记忆（论文分析）
[哼哼]你好。今天我们来看看泰坦在考试时学习记忆的情况。这是 Google 研究的一篇论文，已作为其 Nurips 出版物的一部分推送。所以有一篇关于它的博客文章，人们在社交媒体上发布了关于它的文章，嗯，我不得不承认我……
YouTube — Yannic Kilcher
Publication Catalog
所有文章
All published intelligence, sovereign stack research, ecosystem news, and AI foundation model reviews.
2599
Stories
Исследования1 мин
TiDAR：在扩散中思考，在自回归中讨论（论文分析）
大家好，今天我们来看看 Nvidia 研究人员的 TiDaR 思维扩散讲座和自回归。这是一篇非常酷的论文，因为它有效地证明了您已经观察到我们在自回归大型语言模型期间没有充分利用 GPU……
YouTube — Yannic Kilcher
Исследования1 мин
Transformers, the tech behind LLMs | Deep Learning Chapter 5
The initials GPT stand for Generative Pretrained Transformer. So that first word is straightforward enough, these are bots that generate new text. Pretrained refers to how the model went through a process of learning from a massive amount of data, and the prefix insinuates that…
YouTube — 3Blue1Brown
Исследования1 мин
Attention in transformers, step-by-step | Deep Learning Chapter 6
In the last chapter, you and I started to step through the internal workings of a transformer. This is one of the key pieces of technology inside large language models, and a lot of other tools in the modern wave of AI. It first hit the scene in a now-famous 2017 paper called…
YouTube — 3Blue1Brown
Исследования1 мин
How might LLMs store facts | Deep Learning Chapter 7
If you feed a large language model the phrase, Michael Jordan plays the sport of blank, and you have it predict what comes next, and it correctly predicts basketball, this would suggest that somewhere, inside its hundreds of billions of parameters, it's baked in knowledge about…
YouTube — 3Blue1Brown
Исследования1 мин
Large Language Models explained briefly
[Submit subtitle corrections at criblate.com] Imagine you happen across a short movie script that describes a scene between a person and their AI assistant. The script has what the person asks the AI, but the AI's response has been torn off. Suppose you also have this powerful…
YouTube — 3Blue1Brown
Исследования1 мин
The AI that solved IMO Geometry Problems | Guest video by @Aleph 0
[Submit subtitle corrections at criblate.com] In January 2024, Google DeepMind released an AI model called Alpha Geometry, which could solve geometry problems from the International Mathematical Olympiad, or the IMO. The IMO is the highest level of competitive math contests at…
YouTube — 3Blue1Brown
Исследования1 мин
Reinforcement Learning with Human Feedback (RLHF), Clearly Explained!!!
If you tell me what you like and what you don't like, I'll use that to train my model. Stack Quest. Hello, I'm Josh Starmer and welcome to Stack Quest. Today we're going to talk about reinforcement learning with human feedback and it's going to be clearly explained. This Stack…
YouTube — StatQuest AI
Исследования1 мин
Optimization with Linear Programming (and the Simplex Algorithm), Main Ideas!!!
You start with nothing, then you move. And you keep on moving till you can't get any more, then you're [music] done. StatQuest, bam. Hello, I'm Josh Starmer and welcome to StatQuest. Today, we're going to talk about optimization with linear programming main ideas. Specifically,…
YouTube — StatQuest AI
Исследования1 мин
The Simplex Algorithm, Mathematical Details!!!
We're going to find the optimal solution one row at a time. Stack Quest. Hello, I'm Josh Starmer and welcome to Stat Quest. Today we're going to talk about the simplex algorithm and we're going to go through the mathematical details. This stack quest is brought to you by the…
YouTube — StatQuest AI
Исследования1 мин
优化 ARM CPU 上的专家混合 LLM 推理
大家好，我是来自英特尔印度研究院的 Majid，我很高兴能在今年 10 月的 PyTorch 北美会议上发表演讲。我将讨论使用 VLLM 和 OpenVINO 的专家 LLM 推理的有效组合，以及我们如何在 ARM CPU 上高效运行这些模型。如果…
YouTube — PyTorch Official
Исследования1 мин
PyTorch-Native LLM Serving on TPU: SGLang and vLLM
Hi, I'm G from Google. >> I'm Colin from Meta. >> I'm Angela from Meta. >> We are going to be speaking at the PyTorch conference San Jose about new VLM and SG learn engines built on a new PyTorch native TPU backend torch TPU. >> Rather than build a new serving engine or porting…
YouTube — PyTorch Official
Исследования1 мин
Exchange ideas w/ other devs | on-device AI, inference, voice agents, machine learning compilers...
Hi, I'm Masahiro Hiramoto from Mitsubishi Electric. I'm a member of the program committee for PyTorch conference North America and a PyTorch ambassador. I'm especially looking forward to seeing what people across the community are building with PyTorch and exchanging ideas with…
YouTube — PyTorch Official
Стартапы1 мин
Open Source Inference at Production Scale
Hello everyone, I'm Carlos Costa, a distinction engineer here at IBM Research and co-founder of the LMD project. I'll be heading soon to San Jose for PyTorch Conference North America, where I'll have the privilege of sharing some exciting news about LMD. How the project changed…
YouTube — PyTorch Official
Исследования1 мин
Hugging Face Journal Club: Embarrassingly Simple Self-Distillation Improves Code Generation
Admit, this was actually much more simple than I was imagining, but nonetheless, it's interesting to read. So, we we The title gives it away, right? Embarrassingly >> [laughter] >> In in a previous uh journal club, we read a paper on self-distillation, and in that paper uh there…
YouTube — Hugging Face
Исследования1 мин
Multi-Agent AutoResearch with Open Source Models
Hi, in this video I'm going to walk you through a multi-agent setup of AutoResearch. I use open source models and an open source code harness, OpenCode, and I implement AutoResearch, which is a project by Andre Carpathie where he takes a nano GPT project, an LLM in its simplest…
YouTube — Hugging Face
Исследования1 мин
How to Create an LLM Dataset | FineWeb Overview
Good morning everyone. How's it going today? Today we're going to be talking about the fine web data set. Now, this is a project by Hugging Face and it is a great example of how to gather all the data that you need to train a large language model. This actually is also available…
YouTube — Hugging Face
Исследования1 мин
Hugging Face Journal Club: AsyncOPD and How Stale Can On-Policy Distillation Be?
This paper is doing something if I understand correctly that is very similar to what pipeline RL proposed for reinforcement learning which is where you want to basically have like really decoupled generators from learners from trainers um but you have the additional complexity…
YouTube — Hugging Face
Исследования1 мин
The Hugging Face Hub for Enterprise & Academia
Всем привет. Если вы работаете с ИИ, у вас наверняка уже есть аккаунт на Hugging Face Hub, где удобно демонстрировать свои модели, скачивать модели с открытым кодом, использовать публичные наборы данных и создавать демоверсии своих моделей. Но публичный хаб—это только начало. В…
YouTube — Hugging Face
Исследования1 мин
Hugging Face Journal Club: Kimi K3
Let [clears throat] me think where to start. Okay, so we've got the Kim K3 report and it's quite a beast in terms of details, which is pretty awesome. And the bits that I focused on were really just the the pre- uh sorry, the post-training parts. And even that I wasn't able to…
YouTube — Hugging Face
Исследования1 мин
Hugging Face Journal Club: Scaling Laws for Pre-training & RL
Uh, this paper is actually quite interesting. It dives into something that that has been I guess a question an open question for the Ember communities the Um, we can call them like scaling laws for post-training. And then how is pre-training allocation of compute impact…
YouTube — Hugging Face
Исследования1 мин
Hugging Face Journal Club: Direct On-Policy Distillation
Okay, so this is quite um interesting paper and a bit uh surprising that uh how it works, but uh basically the idea is that uh in post-training, you have usually uh a pipeline where you have a model and then you uh run RL over that model. And ideally that uh post-RL model should…
YouTube — Hugging Face
Исследования1 мин
Optimize, deploy, and benchmark an open-source LLM with vLLM
I'm thrilled to introduce this course on fast and efficient LLM inference, built in partnership with Red Hat, and taught by Sergey Kliger. Deploying open source LLMs efficiently, so they can serve many users at once with low latency and reasonable cost, is challenging. In this…
YouTube — DeepLearning.AI
Стартапы1 мин
Hugging Face Journal Club: Training AI Scientists to Replicate Research
All right. So this paper um is about a model called Faraday um which was trained by this new startup called inherent and what they train the model to do is to do a very narrow part of like scientific replication. So if I understood correctly, basically they want to take a bunch…
YouTube — Hugging Face
Исследования1 мин
Hugging Face Journal Club: AI Research Preference Models
Yeah, I think it's a really cool paper. Um, I think we've talked a bit about this before or like I think it has been a bit in the in the common discourse that like now that implementation is sped up so much, um, like the the value that researchers can provide is more in their…
YouTube — Hugging Face
Исследования1 мин
We shipped 207 WebGPU Kernels for Browser AI
We just shipped more than 200 WebGPU kernels for browser AI and they are all available open source on the Hugging Face Hub. We also shipped a new library called Hugging Face Kernels that allows you to load and to execute all of those kernels. Let's have a look at this little…
YouTube — Hugging Face
Исследования1 мин
Transformers.js v4.3: Structured Output in the browser
We finally released Transformers.js version 4.3 and it's packed with really cool features. Yes, it has been a while since 4.2 and the main reason for that is that we spent a lot of time with the new Hugging Face Canvas library that released last week. But let's have a look at…
YouTube — Hugging Face
Исследования1 мин
Predict next week's weather yourself
Everybody um I am Oritro and uh with me is Emma. So hi. Say hi Emma. >> Hello. >> And uh today what we think we are going to do is a private huddle call uh in hugging face and we are just going to make it live to everybody. And the entire um idea behind this uh call would be to…
YouTube — Hugging Face
Исследования1 мин
HF Buckets: Upload Only What Changed with Xet
Hello everyone. Welcome to this video on stop wasting bandwidth. If you work with large datasets, you're probably re-uploading the same bytes over and over again. But there's a better way, and that's what we're going to discuss in this video. Here's what we're going to cover in…
YouTube — Hugging Face
Исследования1 мин
AI Dev | San Francisco 2026 | 2,599 AI Developers
[music] >> Over the last year, AI Dev Scale from a few hundred people to this one of about [music] 3,000 attendees. So many hallway conversations, great presentations. Someone told me it feels like a music festival to have so many people here chatting, buzzing, talking about…
YouTube — DeepLearning.AI