Keith

@corbtt Curious — does your model’s improved performance on email retrieval come from generating better search queries, or from being able to issue multiple queries when needed?

🚀 Meet ART·E—our open-source RL-trained email research agent that searches your inbox and answers questions more accurately, faster, and cheaper than o3. Let's go deeper on how we built it. 🧵

@jdahanson @dani_avila7 Yes, check out GraphRAG.

@dani_avila7 Would something like this work for complicated documents? Like complex reports or contracts?

“My project became so big that Claude can’t understand it.” Many developers face this problem: LLMs struggle with large codebases. But what if your AI assistant truly understood your entire project? I’ll show you how to build a knowledge graph of your codebase and use CodeGPT to create an AI Copilot that never forgets context. 🧵👇

@janusch_patas @krematas @erikdpeterson Google: multiview 3d reconstruction

@krematas @erikdpeterson Looks good anyway although a bit unsharp. I don't know much about "traditional" dynamic 3D reconstruction. Can you point me to some sota pre-nerf?

So this is what I found from the Olympic broadcasting system. I wonder if this is a NeRF? Seems not to be 3DGS. Any ideas?

@omarsar0 @karpathy Compared to ChatGPT, Siri’s advantage lies in its more comprehensive understanding of the user’s context, which is crucial for an intelligent assistant.

Any thoughts on the new Siri summoning ChatGPT? I understand the importance and opportunity for modularity but I feel that this particular AI experience is unnecessary. Why do I need one assistance to access another assistance and add friction? I really don't get it. Maybe it's just me but curious what others think.

Actually, really liked the Apple Intelligence announcement. It must be a very exciting time at Apple as they layer AI on top of the entire OS. A few of the major themes. Step 1 Multimodal I/O. Enable text/audio/image/video capability, both read and write. These are the native human APIs, so to speak. Step 2 Agentic. Allow all parts of the OS and apps to inter-operate via "function calling"; kernel process LLM that can schedule and coordinate work across them given user queries. Step 3 Frictionless. Fully integrate these features in a highly frictionless, fast, "always on", and contextual way. No going around copy pasting information, prompt engineering, or etc. Adapt the UI accordingly. Step 4 Initiative. Don't perform a task given a prompt, anticipate the prompt, suggest, initiate. Step 5 Delegation hierarchy. Move as much intelligence as you can on device (Apple Silicon very helpful and well-suited), but allow optional dispatch of work to cloud. Step 6 Modularity. Allow the OS to access and support an entire and growing ecosystem of LLMs (e.g. ChatGPT announcement). Step 7 Privacy. <3 We're quickly heading into a world where you can open up your phone and just say stuff. It talks back and it knows you. And it just works. Super exciting and as a user, quite looking forward to it.

@GuillaumeLample For any moderately sized programming project, it is essential to understand the internal libraries or framework mechanisms of the project. Can your LLM-based assistant solve this problem?

Today we are releasing Codestral-22B, our first code model! Codestral is trained on more than 80 programming languages and outperforms the performance of previous code models, including the largest ones. It is available on our API platform, through instruct and fill-in-the-middle endpoints, and can be easily integrated into VScode plugins. You can also use it for free on Le Chat: chat.mistral.ai

"horse made from highly detailed curling swirls, thin glowing multi-colored smoke" by izzybel (ideogram.ai/u/izzybel)

@xuhaiya2483846 支持中文吗

mPLUG-Owl🦉 ranks first on the Multimodal LLM Leaderboard MMBench. Leaderboard: opencompass.org.cn/leaderboard-mu… Github: github.com/X-PLUG/mPLUG-O… Paper: arxiv.org/abs/2304.14178

Looking to hit the next level of growth with a Series A round — but don’t know where to start? 💸 We’ll share everything you need to know in this handy guide, from when to raise to how to find the right investors. #Echobox=1680013576-1" target="_blank" rel="nofollow noopener">sifted.eu/articles/serie…

@Michael_J_Black what's the link of your presentation?

AI for Content Creation, Room 208-210, June 19. My talk starts around 8:55-9:20 CT. Retail Vision, 8:25-9:10 CT, June 19. You can also view them online -- check out the on-line site. How will I do both? The magic of Zoom. I'll be live for questions.

I'm giving two #CVPR2022 workshop talks today. In AI for Content Creation (ai4cc.net) I'll talk about "The Age of Avatars is coming to content creation." In Retail Vision (retailvisionworkshop.github.io) its "Meta-commerce in the Age of Avatars." Details:

@RaritySniperNFT @robofrens @polyheads @Nodeblocks @rarebearnft

@BMANLead btc和eth 只有新高 从来没有所谓的新低

6/在熊市里面，可以选择每周每月定投比特币和以太坊。首先，相对于这个周期的高点，市场已经跌了50%以上，过去的熊市一般会跌去80%-90%，市场可能还没到底，但已经到了低位。其次，按周和按月定投可以分散风险，这是抄底最好的方法。我个人会在每次创新低的时候定投抄底。

《熊市生存指南》 比特币从69000已经跌去超过50%，作为一个13年的老韭菜，上个周期也写过一篇《熊市生存指南》。这个周期，Crypto底层显然已经发生巨大的变化，然而很多的基本规律依然是不变的，那么这个周期的熊市要如何生存，应该采用什么策略？姑且给大家一些老韭菜的经验之谈。

@0xCryptoChan 现在RMB入金还有什么途径？

$BTC 【指标更新】强势突破0轴，星辰大海，已经到来！ 此时不梭，更待何时，冲啊！ #Bitcoin #cryptocurrency

@williamlab 现在RMB入金的途径有哪些？

没有交易所宕机，没有以太坊飙gas，没有大额爆仓，没有插针，没有负费率，没有多空比下降，各个群内还是一片欢声笑语… 这就是本轮瀑布目前的情况。

@Shuyao_Kong please refer to the traditional financial terms investopedia.com/terms/s/smart-…

crypto的smart money到底smart在哪里？

@Rumianian @TheJackForge try @jobsincrypto

@TheJackForge Do you have any tweet about finding a job? I'm so looking for a remote job in React (even an internship).

我妈问我：最近忙活啥呢？忙活元宇宙？ 我：🥲

@Shuyao_Kong Ct是什么

一上CT就发现大家突然都很鸡汤 原来是跌了🥲 不明所以 当然，应该是打工仔本饼，不能理解 大佬们的痛苦

@0xcryptowizard 预测市场不如创造市场

我不是多头，只是追求接近理性/正确的理解市场的本质。 也许经常会犯错，不重要，关键是不断尝试追求逼近正确，市场本就是无法完全预测。 记得刚工作时，还是个intern，和老员工讨论业务，我说，“这件事的本质是” 他看了我一眼，“本质？”眼中全是嘲讽。 若干年过去了，他成为了更老的员工。

Looking to learn all about NFTs? Let us be your guide, w help from @ariannasimpson @cdixon @cuysheffield @dfinzer @Iiterature @jessewldn @katie_haun @kevinchou @lindaxie @muratpak @patrickxrivera @pplpleasr1 @saturnial @sigggnasty @smc90 @zoranbasich et al future.a16z.com