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。

📕 Update: June 16, 2025 Text Input Blocks Released 🎯 We'll be back with even more exciting updates in the coming weeks! ✅ Added 5 Premium Text Input Blocks w / AlignUI Design System

Erşad

14,577 subscribers

21,562 次观看 • 1 年前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

10 条评论

Erşad 的头像

Erşad1 年前

Let's take a look at other blocks ⤵️

Erşad 的头像

Erşad1 年前

➡️ Preview:

Stock.Don 的头像

Stock.Don1 年前

@alignui Why do you need the check in the text fileds?

Erşad 的头像

Erşad1 年前

@alignui Actually, the user can complete the action by pressing the Enter key. The icon is just offered as an extra option. Our goal is to make the components as flexible and customizable as possible.

homeboy 的头像

homeboy1 年前

@alignui i love details 😍

Mohd Sohail 的头像

Mohd Sohail1 年前

@alignui Looks promising!

Swati Awasthi 的头像

Swati Awasthi1 年前

@alignui This kind of polish on input blocks is chef’s kiss 👏 Sleek, clean, and super thoughtful UX—can’t wait to see what’s coming next!

VEERA SRIVASTAVA 的头像

VEERA SRIVASTAVA1 年前

@alignui 🌈 The dashboard here is a masterpiece of elegant design

Sajibur Rahman Sagor 的头像

Sajibur Rahman Sagor1 年前

@alignui Amazing Work

Max Kupriianov 的头像

Max Kupriianov1 年前

@alignui Looks insanely fresh. This is so good!!

相关视频

Say hello to GPT-4o, our new flagship model which can reason across audio, vision, and text in real time: Text and image input rolling out today in API and ChatGPT with voice and video in the coming weeks.

Say hello to GPT-4o, our new flagship model which can reason across audio, vision, and text in real time: Text and image input rolling out today in API and ChatGPT with voice and video in the coming weeks.

OpenAI

22,813,129 次观看 • 2 年前

Ouroboros "short" Leios Protoyping simulation & visualisation work in progress demonstrated by Duncan Coutts. Full P2P network running Leios Visualisation. You can see 100 different nodes and messages flowing between them all, with representation of Input Blocks (Top left, Endorsement Blocks (20 seconds In these increment), Votes (20 seconds in these increment as the pipeline hits a vote stage) You can see the votes continue to burst at roughly 20 seconds, input blocks also increase, the smoothness of message flows to reduce these bursts is something Duncan details they are working on to improve as-well as resource usage. Input Blocks (IB) - These blocks contain the raw transaction data submitted by users to the blockchain. Endorsement Blocks (EB) - These blocks reference multiple Input Blocks and are used by nodes to validate and vote on the readiness of transactions for inclusion in the blockchain. Votes- Nodes cast votes to achieve consensus on the state of the ledger, enhancing the speed of transaction finality. I believe this is completely off-chain so does not incur transaction costs (Someone keep me honest please) Keep an eye on the diffusion latency chart, as Duncan mentions this is very important, this refers more to time taken for propagation of information across the network I believe. This was an extract from a larger demonstration, really exciting, massive credit to the team working on this. It's going to be extremely impactful to Cardano and I will share as much as I can about it, I really like seeing these visualisations especially during a prototyping stage. What I find interesting is that the visualisation vote count is so much it slows down the browser rendering. Great work! web3innovationnerds & Duncan Coutts what a demonstration that was., keep them coming!

Ouroboros "short" Leios Protoyping simulation & visualisation work in progress demonstrated by Duncan Coutts. Full P2P network running Leios Visualisation. You can see 100 different nodes and messages flowing between them all, with representation of Input Blocks (Top left, Endorsement Blocks (20 seconds In these increment), Votes (20 seconds in these increment as the pipeline hits a vote stage) You can see the votes continue to burst at roughly 20 seconds, input blocks also increase, the smoothness of message flows to reduce these bursts is something Duncan details they are working on to improve as-well as resource usage. Input Blocks (IB) - These blocks contain the raw transaction data submitted by users to the blockchain. Endorsement Blocks (EB) - These blocks reference multiple Input Blocks and are used by nodes to validate and vote on the readiness of transactions for inclusion in the blockchain. Votes- Nodes cast votes to achieve consensus on the state of the ledger, enhancing the speed of transaction finality. I believe this is completely off-chain so does not incur transaction costs (Someone keep me honest please) Keep an eye on the diffusion latency chart, as Duncan mentions this is very important, this refers more to time taken for propagation of information across the network I believe. This was an extract from a larger demonstration, really exciting, massive credit to the team working on this. It's going to be extremely impactful to Cardano and I will share as much as I can about it, I really like seeing these visualisations especially during a prototyping stage. What I find interesting is that the visualisation vote count is so much it slows down the browser rendering. Great work! web3innovationnerds & Duncan Coutts what a demonstration that was., keep them coming!

Dave

57,107 次观看 • 1 年前

We've released build 0.1.13 BETA! 540Hz Monitor Giveaway Info Below! This version is set to be the fastest aim trainer yet, with unparalleled input-lag and performance. In the next 48 hours, we'll release a video comparing all the top aim trainers, including KovaaK's, Aimbeast, Aimlabs, and AimingPro. The video will feature full benchmarks and a detailed comparison sheet so you can see the input-lag differences for yourself. (Excel sheet with full video will be included) To celebrate, we're giving away a ROG Swift Pro PG248QP monitor! This monitor, which we purchased specifically for these tests, detects input-lag by connecting your mouse via USB and measuring the time from your click to the monitor's pixel change. We'll post details about how to enter the giveaway tomorrow. Stay tuned for more updates and good luck!

We've released build 0.1.13 BETA! 540Hz Monitor Giveaway Info Below! This version is set to be the fastest aim trainer yet, with unparalleled input-lag and performance. In the next 48 hours, we'll release a video comparing all the top aim trainers, including KovaaK's, Aimbeast, Aimlabs, and AimingPro. The video will feature full benchmarks and a detailed comparison sheet so you can see the input-lag differences for yourself. (Excel sheet with full video will be included) To celebrate, we're giving away a ROG Swift Pro PG248QP monitor! This monitor, which we purchased specifically for these tests, detects input-lag by connecting your mouse via USB and measuring the time from your click to the monitor's pixel change. We'll post details about how to enter the giveaway tomorrow. Stay tuned for more updates and good luck!

EasyAim Trainer

29,140 次观看 • 2 年前

Have you peeped the latest Model Playground update on our serverless inference service? We upgraded our UI so you can test and iterate on models with zero setup. Toggle live updates to see real-time API behavior. Adjust input parameters with full visibility. > Copy-ready requests in Python, TypeScript, cURL, and Gradio > Works across text, vision, image, and audio models > Built on top of the Vercel AI SDK

Have you peeped the latest Model Playground update on our serverless inference service? We upgraded our UI so you can test and iterate on models with zero setup. Toggle live updates to see real-time API behavior. Adjust input parameters with full visibility. > Copy-ready requests in Python, TypeScript, cURL, and Gradio > Works across text, vision, image, and audio models > Built on top of the Vercel AI SDK

Hyperbolic

12,919 次观看 • 1 年前

Switch to 𝕏 Chat 一 Fully Encrypted, No Advertising Hooks "We've evolved the what used to be the old Twitter DM stack, which was unencrypted, basically just text to a fully encrypted messaging system that allows you to do audio and video calls, has all the things you'd want from any messaging app, the disappearing messages, screenshot blocks, like there's a whole all the features that you'd want one in an app. In the next few months, we'll be releasing a standalone chat app."

Switch to 𝕏 Chat 一 Fully Encrypted, No Advertising Hooks "We've evolved the what used to be the old Twitter DM stack, which was unencrypted, basically just text to a fully encrypted messaging system that allows you to do audio and video calls, has all the things you'd want from any messaging app, the disappearing messages, screenshot blocks, like there's a whole all the features that you'd want one in an app. In the next few months, we'll be releasing a standalone chat app."

DogeDesigner

72,442 次观看 • 3 个月前

What's NEXT - FREE Battle Pass?! NEW Revival Device?? More Season Updates | April 18 Update Happy Anniversary! The latest V2.2 update for Farlight 84 will arrive on April 18th. Introducing a bunch of exciting features: a revamped Season system that's longer and more rewarding, Farlight Coins as a new currency packed with even more goodies, a slick new lobby design that's sure to impress, and some fresh in-game tools like the Revival Device and Extended Items. #Farlight84 #Farlight84WhatsNext #Farlight84Anniversary

What's NEXT - FREE Battle Pass?! NEW Revival Device?? More Season Updates | April 18 Update Happy Anniversary! The latest V2.2 update for Farlight 84 will arrive on April 18th. Introducing a bunch of exciting features: a revamped Season system that's longer and more rewarding, Farlight Coins as a new currency packed with even more goodies, a slick new lobby design that's sure to impress, and some fresh in-game tools like the Revival Device and Extended Items. #Farlight84 #Farlight84WhatsNext #Farlight84Anniversary

Farlight 84

59,675 次观看 • 2 年前

New short course: LLMs as Operating Systems: Agent Memory, created with Letta, and taught by its founders Charles Packer and Sarah Wooders. An LLM's input context window has limited space. Using a longer input context also costs more and results in slower processing. So, managing what's stored in this context window is important. In the innovative paper MemGPT: Towards LLMs as Operating Systems, its authors (which include the instructors) proposed using an LLM agent to manage this context window. Their system uses a large persistent memory that stores everything that could be included in the input context, and an agent decides what is actually included. Take the example of building a chatbot that needs to remember what's been said earlier in a conversation (perhaps over many days of interaction with a user). As the conversation's length grows, the memory management agent will move information from the input context to a persistent searchable database; summarize information to keep relevant facts in the input context; and restore relevant conversation elements from further back in time. This allows a chatbot to keep what's currently most relevant in its input context memory to generate the next response. When I read the original MemGPT paper, I thought it was an innovative technique for handling memory for LLMs. The open-source Letta framework, which we'll use in this course, makes MemGPT easy to implement. It adds memory to your LLM agents and gives them transparent long-term memory. In detail, you’ll learn: - How to build an agent that can edit its own limited input context memory, using tools and multi-step reasoning - What is a memory hierarchy (an idea from computer operating systems, which use a cache to speed up memory access), and how these ideas apply to managing the LLM input context (where the input context window is a "cache" storing the most relevant information; and an agent decides what to move in and out of this to/from a larger persistent storage system) - How to implement multi-agent collaboration by letting different agents share blocks of memory This course will give you a sophisticated understanding of memory management for LLMs, which is important for chatbots having long conversations, and for complex agentic workflows. Please sign up here!

New short course: LLMs as Operating Systems: Agent Memory, created with Letta, and taught by its founders Charles Packer and Sarah Wooders. An LLM's input context window has limited space. Using a longer input context also costs more and results in slower processing. So, managing what's stored in this context window is important. In the innovative paper MemGPT: Towards LLMs as Operating Systems, its authors (which include the instructors) proposed using an LLM agent to manage this context window. Their system uses a large persistent memory that stores everything that could be included in the input context, and an agent decides what is actually included. Take the example of building a chatbot that needs to remember what's been said earlier in a conversation (perhaps over many days of interaction with a user). As the conversation's length grows, the memory management agent will move information from the input context to a persistent searchable database; summarize information to keep relevant facts in the input context; and restore relevant conversation elements from further back in time. This allows a chatbot to keep what's currently most relevant in its input context memory to generate the next response. When I read the original MemGPT paper, I thought it was an innovative technique for handling memory for LLMs. The open-source Letta framework, which we'll use in this course, makes MemGPT easy to implement. It adds memory to your LLM agents and gives them transparent long-term memory. In detail, you’ll learn: - How to build an agent that can edit its own limited input context memory, using tools and multi-step reasoning - What is a memory hierarchy (an idea from computer operating systems, which use a cache to speed up memory access), and how these ideas apply to managing the LLM input context (where the input context window is a "cache" storing the most relevant information; and an agent decides what to move in and out of this to/from a larger persistent storage system) - How to implement multi-agent collaboration by letting different agents share blocks of memory This course will give you a sophisticated understanding of memory management for LLMs, which is important for chatbots having long conversations, and for complex agentic workflows. Please sign up here!

Andrew Ng

200,788 次观看 • 1 年前

Thank you all for being with us over the past year. This summer, the adventure continues! We've got even more exciting content coming your way. ☀️The summer update arrives on July 23, bringing new outfits and a variety of limited events for you to enjoy. And that's not the whole story, the new class and an exciting collaboration will be revealed really soon! We'll continue improving the game with new content and a better gameplay experience, and we'll keep fighting alongside with all of our adventurers! #CoA #CrystalofAtlan #1stanniversary #newversion

Thank you all for being with us over the past year. This summer, the adventure continues! We've got even more exciting content coming your way. ☀️The summer update arrives on July 23, bringing new outfits and a variety of limited events for you to enjoy. And that's not the whole story, the new class and an exciting collaboration will be revealed really soon! We'll continue improving the game with new content and a better gameplay experience, and we'll keep fighting alongside with all of our adventurers! #CoA #CrystalofAtlan #1stanniversary #newversion

Crystal_of_Atlan

18,425 次观看 • 12 天前

We have exciting news!💜For our 3D customizable vtuber avatar, the first update (all updates are free! <3) will include an entirely new system for hair customization, allowing for even more options!🩷-^u^- Instead of only having two sections (front and back) now, those have been split, allowing for the bangs and side hair to be selected separately! ✨ All hair groups now have individual color options as well! <3 We will also be including more types of hair in this update! -^u^-

Pastell & Palette🤍 3D Customizable Model!

15,412 次观看 • 8 个月前

Following up on the feedback I gathered a few weeks ago, here's a sneak peek at the new climbing in Hytale! It will be available in next week's pre-release update. Changelog: - Made climbing now solely input-based. - Added rotation around ropes! - No longer constrained to the climbing block (can dismount from the side). - Added pressing "Crouch" to drop vertically without horizontal velocity. - Faster sprint climbing up and down. I got a couple more things I want to do for climbing, like an easier way to catch a ladder when being on top of it, that'll be coming a bit later. I wanted folks to already have a better climbing experience with those features.

Following up on the feedback I gathered a few weeks ago, here's a sneak peek at the new climbing in Hytale! It will be available in next week's pre-release update. Changelog: - Made climbing now solely input-based. - Added rotation around ropes! - No longer constrained to the climbing block (can dismount from the side). - Added pressing "Crouch" to drop vertically without horizontal velocity. - Faster sprint climbing up and down. I got a couple more things I want to do for climbing, like an easier way to catch a ladder when being on top of it, that'll be coming a bit later. I wanted folks to already have a better climbing experience with those features.

Mew

184,972 次观看 • 5 个月前

We’ve been overwhelmed by the feedback on the new Stitch. 🫶 Now, let’s get to building. 😀 Over the coming weeks, we're dropping Starter Ideas to spark your creativity—and want to showcase what you build! 🌟 Starter Idea #1: Watch the new agent and AI-Native Canvas turn a messy doc into a fully styled portfolio. 📝 The Input: Highlight and copy your raw resume text 🎨 The Prompt: "Make me a single-page resume site in 3 fun styles" + paste your resume text ✨ Stitch turns your resume into a text canvas node where it can be referenced or edited ✨ It generates 3 unique Design Systems and layouts 🔷 The Polish: Upload a headshot, and the agent auto-crops and styles it to match the exact aesthetic of each site. We have loved seeing your creations so far. Drop your shared projects in the replies—we're putting together a plan to showcase our favorites!

Stitch by Google

68,096 次观看 • 4 个月前

SYSTEM.DEPLOYMENT_PROTOCOL = UPDATED 🛸 LOCATION.SET: PARASITE_POOL_NODE 🎯 COORDINATES_LOCKED: STRATEGIC_HASHRATE_DEPLOYMENT 🟧 UNIT = ASIC_GENESIS_MINING_CORE 🧰 HARDWARE_STATUS = PRIMED 📦 ALLOCATION_INIT = 200 TH/s (MINIMUM) 🗓️ DEPLOYMENT_DATE = JUNE 14, 2025 🔁 OPERATION_MODE = CONTINUOUS 🌐 POOL.RELATION = STRATEGIC 🤝 ALIGNMENT_STATUS = TRUSTED 📈 HASHRATE.TREND = ASCENDING ⚡ ASIC_GENESIS = EXECUTING 🧠 HASH.WISE = MOVE SMART, MINE STRONG 🔥 LET’S BURN THROUGH BLOCKS TOGETHER 📘 MISSION BRIEF: Starting June 14, the ASIC Genesis unit will begin a strategic mining allocation to Parasite Pool with at least 200 TH/s committed. This is a move to expand decentralized hashrate strength and deepen strategic pool relations. We mine with purpose. We allocate with vision. We build with community. TL;DR: ✅ 200 TH/s → Parasite Pool ✅ June 14 launch ✅ More power, more purpose Transmission live.

SYSTEM.DEPLOYMENT_PROTOCOL = UPDATED 🛸 LOCATION.SET: PARASITE_POOL_NODE 🎯 COORDINATES_LOCKED: STRATEGIC_HASHRATE_DEPLOYMENT 🟧 UNIT = ASIC_GENESIS_MINING_CORE 🧰 HARDWARE_STATUS = PRIMED 📦 ALLOCATION_INIT = 200 TH/s (MINIMUM) 🗓️ DEPLOYMENT_DATE = JUNE 14, 2025 🔁 OPERATION_MODE = CONTINUOUS 🌐 POOL.RELATION = STRATEGIC 🤝 ALIGNMENT_STATUS = TRUSTED 📈 HASHRATE.TREND = ASCENDING ⚡ ASIC_GENESIS = EXECUTING 🧠 HASH.WISE = MOVE SMART, MINE STRONG 🔥 LET’S BURN THROUGH BLOCKS TOGETHER 📘 MISSION BRIEF: Starting June 14, the ASIC Genesis unit will begin a strategic mining allocation to Parasite Pool with at least 200 TH/s committed. This is a move to expand decentralized hashrate strength and deepen strategic pool relations. We mine with purpose. We allocate with vision. We build with community. TL;DR: ✅ 200 TH/s → Parasite Pool ✅ June 14 launch ✅ More power, more purpose Transmission live.

ASIC Genesis

11,354 次观看 • 1 年前

🚨 Paper Alert 🚨 ➡️Paper Title: Articulate3D: Zero-Shot Text-Driven 3D Object Posing 🌟Few pointers from the paper 🎯Authors of this paper proposed a training-free method, “Articulate3D”, to pose a 3D asset through language control. 🎯Despite advances in vision and language models, this task remains surprisingly challenging. 🎯To achieve this goal, they decomposed the problem into two steps. 🎯They modified a powerful image-generator to create target images conditioned on the input image and a text instruction. 🎯They then align the mesh to the target images through a multi-view pose optimisation step. 🎯 In detail, they introduced a self-attention rewiring mechanism (RSActrl) that decouples the source structure from pose within an image generative model, allowing it to maintain a consistent structure across varying poses. 🎯They observed that differentiable rendering is an unreliable signal for articulation optimisation; instead, they used keypoints to establish correspondences between input and target images. 🎯The effectiveness of Articulate3D is demonstrated across a diverse range of 3D objects and free-form text prompts, successfully manipulating poses while maintaining the original identity of the mesh. 🎯Quantitative evaluations and a comparative user study, in which their method was preferred over 85% of the time, confirm its superiority over existing approaches. 🏢Organization: University of Oxford , Google DeepMind 🧙Paper Authors: Oishi Deb, Anjun Hu, Ashkan Khakzar, Philip Torr, Christian Rupprecht 📝 Read the Full Paper here: 🗂️ Project Page: 🎥 Be sure to watch the attached Demo Video - Sound on 🔊🔊 Find this Valuable 💎 ? ♻️QT and teach your network something new Follow me 👣, naveen manwani , for the latest updates on Tech and AI-related news, insightful research papers, and exciting announcements.

🚨 Paper Alert 🚨 ➡️Paper Title: Articulate3D: Zero-Shot Text-Driven 3D Object Posing 🌟Few pointers from the paper 🎯Authors of this paper proposed a training-free method, “Articulate3D”, to pose a 3D asset through language control. 🎯Despite advances in vision and language models, this task remains surprisingly challenging. 🎯To achieve this goal, they decomposed the problem into two steps. 🎯They modified a powerful image-generator to create target images conditioned on the input image and a text instruction. 🎯They then align the mesh to the target images through a multi-view pose optimisation step. 🎯 In detail, they introduced a self-attention rewiring mechanism (RSActrl) that decouples the source structure from pose within an image generative model, allowing it to maintain a consistent structure across varying poses. 🎯They observed that differentiable rendering is an unreliable signal for articulation optimisation; instead, they used keypoints to establish correspondences between input and target images. 🎯The effectiveness of Articulate3D is demonstrated across a diverse range of 3D objects and free-form text prompts, successfully manipulating poses while maintaining the original identity of the mesh. 🎯Quantitative evaluations and a comparative user study, in which their method was preferred over 85% of the time, confirm its superiority over existing approaches. 🏢Organization: University of Oxford , Google DeepMind 🧙Paper Authors: Oishi Deb, Anjun Hu, Ashkan Khakzar, Philip Torr, Christian Rupprecht 📝 Read the Full Paper here: 🗂️ Project Page: 🎥 Be sure to watch the attached Demo Video - Sound on 🔊🔊 Find this Valuable 💎 ? ♻️QT and teach your network something new Follow me 👣, naveen manwani , for the latest updates on Tech and AI-related news, insightful research papers, and exciting announcements.

naveen manwani

14,334 次观看 • 11 个月前

⚙️Update Quick Video Review: GameSir Connect V1.14.4⚙️ GameSir Connect has just been updated to version V1.14.4. Over the past few updates, we have introduced many practical features to enhance your experience. Here is the video detailing all the highlights of the recent updates. We highly recommend that players visit the Microsoft Store to download the latest version from here: For details on the updates, please refer to the text below. ✅Feature 1 - Professional Tips and Explanation for Controller Settings. You can now hover your mouse over the ⓘ icon to access detailed explanations of controller terminology. These tips have been verified by the GameSir product team to ensure you can quickly, conveniently, and accurately understand the information you need. ✅Feature 2 - Enhanced Stick Customization: Expanded Curve & Calibration Settings Now, you can adjust the stick center point to fix any potential center offset after the controller has already been calibrated. In addition, you can now achieve more precise fine-tuning of the stick input curve by manually entering the coordinates for each point within a 100x100 coordinate system. Please be aware that when Zero Deadzone is enabled, it is normal for the center point to not return to absolute zero or for the data to fluctuate repeatedly. ✅Feature 3 - Support for in-app software updates and halved input latency We've now set up a dedicated server for GameSir Connect software updates. Now, you can update directly within the software. And through optimizations, the end-to-end latency for both the G7 Pro 8K PC and Tarantula 8K PC has now reached 0.25ms with the latest firmware. And only by using the latest software will you be able to receive the latest firmware updates.

⚙️Update Quick Video Review: GameSir Connect V1.14.4⚙️ GameSir Connect has just been updated to version V1.14.4. Over the past few updates, we have introduced many practical features to enhance your experience. Here is the video detailing all the highlights of the recent updates. We highly recommend that players visit the Microsoft Store to download the latest version from here: For details on the updates, please refer to the text below. ✅Feature 1 - Professional Tips and Explanation for Controller Settings. You can now hover your mouse over the ⓘ icon to access detailed explanations of controller terminology. These tips have been verified by the GameSir product team to ensure you can quickly, conveniently, and accurately understand the information you need. ✅Feature 2 - Enhanced Stick Customization: Expanded Curve & Calibration Settings Now, you can adjust the stick center point to fix any potential center offset after the controller has already been calibrated. In addition, you can now achieve more precise fine-tuning of the stick input curve by manually entering the coordinates for each point within a 100x100 coordinate system. Please be aware that when Zero Deadzone is enabled, it is normal for the center point to not return to absolute zero or for the data to fluctuate repeatedly. ✅Feature 3 - Support for in-app software updates and halved input latency We've now set up a dedicated server for GameSir Connect software updates. Now, you can update directly within the software. And through optimizations, the end-to-end latency for both the G7 Pro 8K PC and Tarantula 8K PC has now reached 0.25ms with the latest firmware. And only by using the latest software will you be able to receive the latest firmware updates.

GameSir

27,951 次观看 • 14 天前

If you're building a PDF RAG pipeline: Should you be using OCR and 𝘁𝗲𝘅𝘁-𝗯𝗮𝘀𝗲𝗱 𝗿𝗲𝘁𝗿𝗶𝗲𝘃𝗮𝗹 methods, or just 𝗲𝗺𝗯𝗲𝗱 𝗶𝗺𝗮𝗴𝗲𝘀 𝗱𝗶𝗿𝗲𝗰𝘁𝗹𝘆 using late interaction models? This paper says the answer might actually be 𝘣𝘰𝘵𝘩. My colleagues at Weaviate released IRPAPERS, a benchmark comparing 𝗶𝗺𝗮𝗴𝗲-𝗯𝗮𝘀𝗲𝗱 and 𝘁𝗲𝘅𝘁-𝗯𝗮𝘀𝗲𝗱 retrieval over 3,230 pages from 166 scientific papers. The setup: Take the same PDFs and process them two ways. For text, run OCR with GPT-4.1 and embed with Arctic 2.0 + BM25 hybrid search. For images, embed raw page images with ColModernVBERT multi-vector embeddings. Test both on 180 needle-in-the-haystack questions. 𝗧𝗵𝗲 𝗿𝗲𝘀𝘂𝗹𝘁𝘀: Text edges out images at the top rank: 46% vs 43% Recall@1 But images match or exceed text at deeper recall: 93% vs 91% Recall@20 But text and image based methods actually fail on 𝘥𝘪𝘧𝘧𝘦𝘳𝘦𝘯𝘁 𝘲𝘶𝘦𝘳𝘪𝘦𝘴. At Recall@1: • 22 queries succeed with text but fail with images • 18 queries succeed with images but fail with text This complementarity is what makes 𝗠𝘂𝗹𝘁𝗶𝗺𝗼𝗱𝗮𝗹 𝗛𝘆𝗯𝗿𝗶𝗱 𝗦𝗲𝗮𝗿𝗰𝗵 work. By fusing scores from both text and image retrieval, they achieved: • 49% Recall@1 (beating either modality alone) • 81% Recall@5 • 95% Recall@20 More in the video below 🔽 Dataset: Paper: Code:

If you're building a PDF RAG pipeline: Should you be using OCR and 𝘁𝗲𝘅𝘁-𝗯𝗮𝘀𝗲𝗱 𝗿𝗲𝘁𝗿𝗶𝗲𝘃𝗮𝗹 methods, or just 𝗲𝗺𝗯𝗲𝗱 𝗶𝗺𝗮𝗴𝗲𝘀 𝗱𝗶𝗿𝗲𝗰𝘁𝗹𝘆 using late interaction models? This paper says the answer might actually be 𝘣𝘰𝘵𝘩. My colleagues at Weaviate released IRPAPERS, a benchmark comparing 𝗶𝗺𝗮𝗴𝗲-𝗯𝗮𝘀𝗲𝗱 and 𝘁𝗲𝘅𝘁-𝗯𝗮𝘀𝗲𝗱 retrieval over 3,230 pages from 166 scientific papers. The setup: Take the same PDFs and process them two ways. For text, run OCR with GPT-4.1 and embed with Arctic 2.0 + BM25 hybrid search. For images, embed raw page images with ColModernVBERT multi-vector embeddings. Test both on 180 needle-in-the-haystack questions. 𝗧𝗵𝗲 𝗿𝗲𝘀𝘂𝗹𝘁𝘀: Text edges out images at the top rank: 46% vs 43% Recall@1 But images match or exceed text at deeper recall: 93% vs 91% Recall@20 But text and image based methods actually fail on 𝘥𝘪𝘧𝘧𝘦𝘳𝘦𝘯𝘁 𝘲𝘶𝘦𝘳𝘪𝘦𝘴. At Recall@1: • 22 queries succeed with text but fail with images • 18 queries succeed with images but fail with text This complementarity is what makes 𝗠𝘂𝗹𝘁𝗶𝗺𝗼𝗱𝗮𝗹 𝗛𝘆𝗯𝗿𝗶𝗱 𝗦𝗲𝗮𝗿𝗰𝗵 work. By fusing scores from both text and image retrieval, they achieved: • 49% Recall@1 (beating either modality alone) • 81% Recall@5 • 95% Recall@20 More in the video below 🔽 Dataset: Paper: Code:

Victoria Slocum

43,996 次观看 • 4 个月前

🚀 Exciting progress in the #EndlessRangerAwakening! Check out the latest updates we've made to enhance your gaming experience: **Game UI Design:** ✅ Feedback interface integrated. ✅ UI texts have been redesigned. ✅ Interface elements within the arena can now be hidden. ✅ New attack pet, Felinix, added. ✅ Stun icon integrated. ✅ Message interface history clear button added. ✅ Reward collection pop-ups integrated. ✅ Custom code entry interface for closed beta integrated. ✅ Space Station arena corridors widened and lighting revised. ✅ Actiswap and Mission helper NPCs revised and added to the game. **Game Development:** ✅ Transitioned to a new server for the closed beta. ✅ Dracarys Boss damage revised and crowd control added. ✅ Messaging filter integrated to block profanity, slang, and link sharing. ✅ Anti-spam penalty system integrated into messaging. ✅ Issues with character skill direction fixed. ✅ One-time use code system set up for closed beta PC. ✅ FPS camera mode shaking issue fixed. ✅ Loading screen freezing issues resolved. 🔗 We've moved to a new server for the closed beta, revamped Dracarys Boss damage, and added crowd control. Plus, we've integrated filters to block profanity and spam in messaging, and resolved issues with character skill directions. 📈 We're thrilled with these enhancements and can't wait to launch our Blockchain PlayToEarn MMORPG, taking your game experience to the next level! 👀 Stay tuned for more updates and upcoming game releases. Join us on this journey to the beta and beyond! #BlockchainGaming #PlayToEarn #GameDev

🚀 Exciting progress in the #EndlessRangerAwakening! Check out the latest updates we've made to enhance your gaming experience: Game UI Design: ✅ Feedback interface integrated. ✅ UI texts have been redesigned. ✅ Interface elements within the arena can now be hidden. ✅ New attack pet, Felinix, added. ✅ Stun icon integrated. ✅ Message interface history clear button added. ✅ Reward collection pop-ups integrated. ✅ Custom code entry interface for closed beta integrated. ✅ Space Station arena corridors widened and lighting revised. ✅ Actiswap and Mission helper NPCs revised and added to the game. Game Development: ✅ Transitioned to a new server for the closed beta. ✅ Dracarys Boss damage revised and crowd control added. ✅ Messaging filter integrated to block profanity, slang, and link sharing. ✅ Anti-spam penalty system integrated into messaging. ✅ Issues with character skill direction fixed. ✅ One-time use code system set up for closed beta PC. ✅ FPS camera mode shaking issue fixed. ✅ Loading screen freezing issues resolved. 🔗 We've moved to a new server for the closed beta, revamped Dracarys Boss damage, and added crowd control. Plus, we've integrated filters to block profanity and spam in messaging, and resolved issues with character skill directions. 📈 We're thrilled with these enhancements and can't wait to launch our Blockchain PlayToEarn MMORPG, taking your game experience to the next level! 👀 Stay tuned for more updates and upcoming game releases. Join us on this journey to the beta and beyond! #BlockchainGaming #PlayToEarn #GameDev

BRN Metaverse

66,682 次观看 • 1 年前

The IGNITION season is here, and we've got a ton of exciting new content waiting for you! In this video, we'll give you a closer look at everything coming in the new season, including: ✅ New Northridge map ✅ New bosses and new challenges ✅ Fresh new weapons like the MCX and AK12 ✅ New modes: 4v4 Team Death Match, Solo Ops, and More… ✅ Improved system features for smoother gameplay ✅ Balance changes based on community feedback Get ready for intense action and all-new tactical gameplay. Watch the video and tell us which update you're most excited for in the comments! #arenabreakoutinfinite #ArenaBreakoutPC #ABINewSeason

The IGNITION season is here, and we've got a ton of exciting new content waiting for you! In this video, we'll give you a closer look at everything coming in the new season, including: ✅ New Northridge map ✅ New bosses and new challenges ✅ Fresh new weapons like the MCX and AK12 ✅ New modes: 4v4 Team Death Match, Solo Ops, and More… ✅ Improved system features for smoother gameplay ✅ Balance changes based on community feedback Get ready for intense action and all-new tactical gameplay. Watch the video and tell us which update you're most excited for in the comments! #arenabreakoutinfinite #ArenaBreakoutPC #ABINewSeason

Arena Breakout Infinite

252,339 次观看 • 1 年前

Exciting Update: The freshly redesigned Reactor and our new website are here! What's new: - Upgraded Reactor & Website UI with New Design - Focused Matrix Pages (MVP on the way🤫) - More Sections for B2B Opportunities (get in contact for Reactor API access or Reactor ($ARC on Virtuals) integration) We’re putting the finishing touches on a few features and pages, which will be fully publicly available in the next couple of days. With this update, including the Reactor enhancement and Matrix development, we expect new opportunities and high-level collaborations coming our way in the next months.🤝 Head over to and and test yourself! For more updates, follow ARC and our CEO Tj Dunham.

Exciting Update: The freshly redesigned Reactor and our new website are here! What's new: - Upgraded Reactor & Website UI with New Design - Focused Matrix Pages (MVP on the way🤫) - More Sections for B2B Opportunities (get in contact for Reactor API access or Reactor ($ARC on Virtuals) integration) We’re putting the finishing touches on a few features and pages, which will be fully publicly available in the next couple of days. With this update, including the Reactor enhancement and Matrix development, we expect new opportunities and high-level collaborations coming our way in the next months.🤝 Head over to and and test yourself! For more updates, follow ARC and our CEO Tj Dunham.

ARC

28,955 次观看 • 1 年前

Never underestimate the speed at which visionOS is evolving. Just three weeks later, with the latest beta, the Logitech Muse now supports almost any app and can be used for input across the entire system. I tested it with my favorite drawing app, TouchDesk, and it’s incredibly smooth – it feels just like using an Apple Pencil since you don’t even have to press a button to draw. You just touch the desk. There’s no lag, and it even provides subtle haptic feedback while drawing. The last time I reviewed it, I wasn’t entirely sure I could recommend it, but with the new system-wide support coming to visionOS, I think it’s now a must-have for every Vision Pro enthusiast.

Never underestimate the speed at which visionOS is evolving. Just three weeks later, with the latest beta, the Logitech Muse now supports almost any app and can be used for input across the entire system. I tested it with my favorite drawing app, TouchDesk, and it’s incredibly smooth – it feels just like using an Apple Pencil since you don’t even have to press a button to draw. You just touch the desk. There’s no lag, and it even provides subtle haptic feedback while drawing. The last time I reviewed it, I wasn’t entirely sure I could recommend it, but with the new system-wide support coming to visionOS, I think it’s now a must-have for every Vision Pro enthusiast.

Phil Traut ᯅ

279,082 次观看 • 8 个月前