Loading video...

Video Failed to Load

Go Home

The web was never meant to be flattened into text. Yet most web RAG systems start by parsing HTML --- a complex and lossy process. 🔥 Introducing PixelRAG: the first RAG system that retrieves and reads 30M+ web pages as pixels. Instead of extracting text, PixelRAG retrieves screenshots and...

YichuanM's profile picture

Yichuan Wang

2,264 subscribers

89,492 views • 1 month ago •via X (Twitter)

EducationScience & Technology
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Finally, a RAG solution that works with complex documents! Real-world documents can be messy, filled with text, tables, images, and intricate flow charts. Traditional parsing and chunking methods struggle to handle these. So, what’s the solution? We need smart techniques that can intuitively chunk relevant context and understand what’s inside each chunk, whether it's text, images, or diagrams. In this video, I’ll walk you through a breakthrough technique for extracting structured information from complex documents. It's unlike any other technique you've seen before.✨ It takes any unstructured (text, tables, images, flow-charts) input and parses it into a JSON format that LLMs can easily process. I used EyeLevel.AI's GroundX platform for this – a powerful tool that allows you to build a RAG application in just 3 steps. It also comes with a nice Python SDK and can be easily deployed on-premise (K8s cluster)! Try it yourself:
6:45

Finally, a RAG solution that works with complex documents! Real-world documents can be messy, filled with text, tables, images, and intricate flow charts. Traditional parsing and chunking methods struggle to handle these. So, what’s the solution? We need smart techniques that can intuitively chunk relevant context and understand what’s inside each chunk, whether it's text, images, or diagrams. In this video, I’ll walk you through a breakthrough technique for extracting structured information from complex documents. It's unlike any other technique you've seen before.✨ It takes any unstructured (text, tables, images, flow-charts) input and parses it into a JSON format that LLMs can easily process. I used EyeLevel.AI's GroundX platform for this – a powerful tool that allows you to build a RAG application in just 3 steps. It also comes with a nice Python SDK and can be easily deployed on-premise (K8s cluster)! Try it yourself:

Akshay 🚀

102,660 views • 1 year ago

Working on multimodal instruction tuning and finding it hard to scale? Building Web/GUI agents but data is too narrow? Introducing 🚀MultiUI: 7.3M multimodal instructions from 1M webpage UIs, offering diverse data to boost text-rich visual understanding. Key takeaways: 🌟WebUI-trained models show major gains in visual web understanding and agent tasks. 💻 🌟Models also generalize well to non-UI tasks like DocVQA/OCR. 📄 How it works: We generate multimodal instructions with a text LLM using structured text from webpage accessibility trees. We then pair them with UI screenshots, to train multimodal models. Homepage: Paper: Dataset: Model: Congrats to the student lead Junpeng Liu and the team Tianyue Ou Yifan Song Yuxiao Qu Chenyan Xiong Wenhu Chen Graham Neubig ! More details are in the following threads ⬇️
1:19

Working on multimodal instruction tuning and finding it hard to scale? Building Web/GUI agents but data is too narrow? Introducing 🚀MultiUI: 7.3M multimodal instructions from 1M webpage UIs, offering diverse data to boost text-rich visual understanding. Key takeaways: 🌟WebUI-trained models show major gains in visual web understanding and agent tasks. 💻 🌟Models also generalize well to non-UI tasks like DocVQA/OCR. 📄 How it works: We generate multimodal instructions with a text LLM using structured text from webpage accessibility trees. We then pair them with UI screenshots, to train multimodal models. Homepage: Paper: Dataset: Model: Congrats to the student lead Junpeng Liu and the team Tianyue Ou Yifan Song Yuxiao Qu Chenyan Xiong Wenhu Chen Graham Neubig ! More details are in the following threads ⬇️

Xiang Yue

57,680 views • 1 year ago

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 views • 3 years ago

Introducing RAGs v3 🌟: Build a RAG bot that can also search the web 🌐, to find answers that aren’t immediately in its corpus. Do this all in natural language, not code! 💬 Get a better experience than ChatGPT + Bing ⚡️ We used our integration with Metaphor - a search engine designed for LLMs - to pull in relevant text from the internet. Additionally: - you can now view the tools that the agent is using. - note: web search is only accessible for our OpenAI agent Check out the video below, and get started with the repo today. Repo:
1:17

Introducing RAGs v3 🌟: Build a RAG bot that can also search the web 🌐, to find answers that aren’t immediately in its corpus. Do this all in natural language, not code! 💬 Get a better experience than ChatGPT + Bing ⚡️ We used our integration with Metaphor - a search engine designed for LLMs - to pull in relevant text from the internet. Additionally: - you can now view the tools that the agent is using. - note: web search is only accessible for our OpenAI agent Check out the video below, and get started with the repo today. Repo:

LlamaIndex 🦙

76,366 views • 2 years ago

PDF parsing is still painful because LLMs reorder text in complex layouts, break tables across pages, and fail on graphs or images. 💡Testing the new open-source OCRFlux model, and here the results are really good for a change. So OCRFlux is a multimodal, LLM based toolkit for converting PDFs and images into clean, readable, plain Markdown text. Because the underlying VLM is only 3B param, it runs even on a 3090 GPU. The model is available on Hugging Face . The engine that powers the OCRFlux, teaches the model to rebuild every page and then stitch fragments across pages into one clean Markdown file. It bundles one vision language model with 3B parameters that was fine-tuned from Qwen 2.5-VL-3B-Instruct for both page parsing and cross-page merging. OCRFlux reads raw page images and, guided by task prompts, outputs Markdown for each page and merges split elements across pages. The evaluation shows Edit Distance Similarity (EDS) 0.967 and cross‑page table Tree Edit Distance 0.950, so the parser is both accurate and layout aware. How it works while parsing each page - Convert into text with a natural reading order, even in the presence of multi-column layouts, figures, and insets - Support for complicated tables and equations - Automatically removes headers and footers Cross-page table/paragraph merging - Cross-page table merging - Cross-page paragraph merging A compact vision‑language models can beat bigger models once cross‑page context is added. 🧵 1/n Read on 👇
0:44

PDF parsing is still painful because LLMs reorder text in complex layouts, break tables across pages, and fail on graphs or images. 💡Testing the new open-source OCRFlux model, and here the results are really good for a change. So OCRFlux is a multimodal, LLM based toolkit for converting PDFs and images into clean, readable, plain Markdown text. Because the underlying VLM is only 3B param, it runs even on a 3090 GPU. The model is available on Hugging Face . The engine that powers the OCRFlux, teaches the model to rebuild every page and then stitch fragments across pages into one clean Markdown file. It bundles one vision language model with 3B parameters that was fine-tuned from Qwen 2.5-VL-3B-Instruct for both page parsing and cross-page merging. OCRFlux reads raw page images and, guided by task prompts, outputs Markdown for each page and merges split elements across pages. The evaluation shows Edit Distance Similarity (EDS) 0.967 and cross‑page table Tree Edit Distance 0.950, so the parser is both accurate and layout aware. How it works while parsing each page - Convert into text with a natural reading order, even in the presence of multi-column layouts, figures, and insets - Support for complicated tables and equations - Automatically removes headers and footers Cross-page table/paragraph merging - Cross-page table merging - Cross-page paragraph merging A compact vision‑language models can beat bigger models once cross‑page context is added. 🧵 1/n Read on 👇

Rohan Paul

149,292 views • 1 year ago

We’re excited to officially launch LlamaParse, the first genAI-native document parsing solution. Not only is it better at parsing out images/tables/charts 📊📈 than virtually every other parser, it is now steerable through natural language instructions - output the document in whatever format you desire! It is also the only parsing solution that seamlessly allows you to build accurate RAG over complex documents, free of hallucinations 🔥 We launched it in private preview a few weeks ago and hit 2k users, 1M total PDF pages parsed. And now it’s better than ever. LlamaParse contains the following killer features: ✅ SOTA table/chart extraction ✅ Seamless integration with LlamaIndex 🦙 advanced RAG/agents ✅✨ Natural language Parsing Instructions ✅✨JSON mode and image extraction ✅✨Support for ~10 document types (.pdf, .pptx, .docx, .xml) and more Our pricing is simple: 1k free per day, and additional pages at 0.3c a page, or $3 for 1k pages. If you want advanced document RAG and/or private deployments, come get in touch with us to chat about LlamaCloud. Check out our full blog post here: LlamaParse client repo: Signup at 🦙☁️: Come talk to us:
5:07

We’re excited to officially launch LlamaParse, the first genAI-native document parsing solution. Not only is it better at parsing out images/tables/charts 📊📈 than virtually every other parser, it is now steerable through natural language instructions - output the document in whatever format you desire! It is also the only parsing solution that seamlessly allows you to build accurate RAG over complex documents, free of hallucinations 🔥 We launched it in private preview a few weeks ago and hit 2k users, 1M total PDF pages parsed. And now it’s better than ever. LlamaParse contains the following killer features: ✅ SOTA table/chart extraction ✅ Seamless integration with LlamaIndex 🦙 advanced RAG/agents ✅✨ Natural language Parsing Instructions ✅✨JSON mode and image extraction ✅✨Support for ~10 document types (.pdf, .pptx, .docx, .xml) and more Our pricing is simple: 1k free per day, and additional pages at 0.3c a page, or $3 for 1k pages. If you want advanced document RAG and/or private deployments, come get in touch with us to chat about LlamaCloud. Check out our full blog post here: LlamaParse client repo: Signup at 🦙☁️: Come talk to us:

LlamaIndex 🦙

143,136 views • 2 years ago

Yann LeCun explains that large language models are trained on about 30 trillion words, representing nearly all public internet text. He says it would take a human over 500,000 years to read that much. But a 4-year-old child sees just as much visual data in their first few years of life. This shows how much richer and more complex real-world experience is compared to reading text. Training on the web is huge but it still doesn’t match what a child learns just by living.
2:41

Yann LeCun explains that large language models are trained on about 30 trillion words, representing nearly all public internet text. He says it would take a human over 500,000 years to read that much. But a 4-year-old child sees just as much visual data in their first few years of life. This shows how much richer and more complex real-world experience is compared to reading text. Training on the web is huge but it still doesn’t match what a child learns just by living.

Wes Roth

294,390 views • 7 months ago

Marc Andreessen reveals one of the wildest sliding doors moments in tech history. In 1994, he almost pulled the trigger on building an online gaming service for the Nintendo 64. He was essentially planning to build Xbox Live or the PlayStation Network years before they existed. They decided it was too early and walked away. But the real reason they pivoted? The internet just kept growing. Instead of revolutionizing console gaming, his team focused on a college project called Mosaic. At the time, the internet was not viewed as a consumer phenomenon at all. The web existed as a nascent concept, but it was limited to clunky, text-based terminals. There were no graphics, no point-and-click capabilities, no scripting languages, and absolutely no security. It lacked the actual capabilities that make a browser a useful tool for the public. Andreessen and his group at Illinois completely changed the trajectory of technology. They built the first widely used web browser with graphics, turning that early text-based concept into the full, capable software experience we know today. To support it, they also built the first mainstream web server that had everything everyday people needed to actually get online. It is incredible to look back and realize that the foundation of the modern visual web was just a college project that almost took a backseat to a Nintendo 64 online network. Ft: Marc Andreessen 🇺🇸 David Senra
1:39

Marc Andreessen reveals one of the wildest sliding doors moments in tech history. In 1994, he almost pulled the trigger on building an online gaming service for the Nintendo 64. He was essentially planning to build Xbox Live or the PlayStation Network years before they existed. They decided it was too early and walked away. But the real reason they pivoted? The internet just kept growing. Instead of revolutionizing console gaming, his team focused on a college project called Mosaic. At the time, the internet was not viewed as a consumer phenomenon at all. The web existed as a nascent concept, but it was limited to clunky, text-based terminals. There were no graphics, no point-and-click capabilities, no scripting languages, and absolutely no security. It lacked the actual capabilities that make a browser a useful tool for the public. Andreessen and his group at Illinois completely changed the trajectory of technology. They built the first widely used web browser with graphics, turning that early text-based concept into the full, capable software experience we know today. To support it, they also built the first mainstream web server that had everything everyday people needed to actually get online. It is incredible to look back and realize that the foundation of the modern visual web was just a college project that almost took a backseat to a Nintendo 64 online network. Ft: Marc Andreessen 🇺🇸 David Senra

Ian Miles Cheong

19,526 views • 2 months ago

Announcing a new Coursera course: Retrieval Augmented Generation (RAG) You'll learn to build high performance, production-ready RAG systems in this hands-on, in-depth course created by and taught by Zain, experienced AI and ML engineer, researcher, and educator. RAG is a critical component today of many LLM-based applications in customer support, internal company Q&A systems, even many of the leading chatbots that use web search to answer your questions. This course teaches you in-depth how to make RAG work well. LLMs can produce generic or outdated responses, especially when asked specialized questions not covered in its training data. RAG is the most widely used technique for addressing this. It brings in data from new data sources, such as internal documents or recent news, to give the LLM the relevant context to private, recent, or specialized information. This lets it generate more grounded and accurate responses. In this course, you’ll learn to design and implement every part of a RAG system, from retrievers to vector databases to generation to evals. You’ll learn about the fundamental principles behind RAG and how to optimize it at both the component and whole-system levels. As AI evolves, RAG is evolving too. New models can handle longer context windows, reason more effectively, and can be parts of complex agentic workflows. One exciting growth area is Agentic RAG, in which an AI agent at runtime (rather than it being hardcoded at development time) autonomously decides what data to retrieve, and when/how to go deeper. Even with this evolution, access to high-quality data at runtime is essential, which is why RAG is a key part of so many applications. You'll learn via hands-on experiences to: - Build a RAG system with retrieval and prompt augmentation - Compare retrieval methods like BM25, semantic search, and Reciprocal Rank Fusion - Chunk, index, and retrieve documents using a Weaviate vector database and a news dataset - Develop a chatbot, using open-source LLMs hosted by Together AI, for a fictional store that answers product and FAQ questions - Use evals to drive improving reliability, and incorporate multi-modal data RAG is an important foundational technique. Become good at it through this course! Please sign up here:
1:39

Announcing a new Coursera course: Retrieval Augmented Generation (RAG) You'll learn to build high performance, production-ready RAG systems in this hands-on, in-depth course created by and taught by Zain, experienced AI and ML engineer, researcher, and educator. RAG is a critical component today of many LLM-based applications in customer support, internal company Q&A systems, even many of the leading chatbots that use web search to answer your questions. This course teaches you in-depth how to make RAG work well. LLMs can produce generic or outdated responses, especially when asked specialized questions not covered in its training data. RAG is the most widely used technique for addressing this. It brings in data from new data sources, such as internal documents or recent news, to give the LLM the relevant context to private, recent, or specialized information. This lets it generate more grounded and accurate responses. In this course, you’ll learn to design and implement every part of a RAG system, from retrievers to vector databases to generation to evals. You’ll learn about the fundamental principles behind RAG and how to optimize it at both the component and whole-system levels. As AI evolves, RAG is evolving too. New models can handle longer context windows, reason more effectively, and can be parts of complex agentic workflows. One exciting growth area is Agentic RAG, in which an AI agent at runtime (rather than it being hardcoded at development time) autonomously decides what data to retrieve, and when/how to go deeper. Even with this evolution, access to high-quality data at runtime is essential, which is why RAG is a key part of so many applications. You'll learn via hands-on experiences to: - Build a RAG system with retrieval and prompt augmentation - Compare retrieval methods like BM25, semantic search, and Reciprocal Rank Fusion - Chunk, index, and retrieve documents using a Weaviate vector database and a news dataset - Develop a chatbot, using open-source LLMs hosted by Together AI, for a fictional store that answers product and FAQ questions - Use evals to drive improving reliability, and incorporate multi-modal data RAG is an important foundational technique. Become good at it through this course! Please sign up here:

Andrew Ng

124,458 views • 1 year ago

New Short Course: Building AI Browser Agents! Learn how to build AI agents that interact and take actions on websites in this course, created in partnership with and taught by and @namangarg0, Co-founders of AGI Inc. AI browser agents can log into websites, fill out forms, click through web pages, or even place orders online for you. They use both visual information, like screenshots, and structural data, like the HTML or Document Object Model (DOM) of a web page, to reason and take action. With the complexity of webpages and multiple possible actions at each step, it can be challenging for an AI browser agent to complete an assigned task. Because these agents run long action sequences, a single error—like clicking the wrong button or misreading a field—can lead to unexpected outcomes or errors that compound over time. In this course, you'll understand how autonomous web agents work, their current limitations, and how AgentQ enables them to improve through self-correction. In detail, you'll: - Learn what web agents are, how they automate tasks online, their architecture, key components, limitations, and an overview of their decision-making strategies. - Build a web agent that can scrape website and return course recommendations in a structured output format. - Build an autonomous web agent that can execute multiple tasks, such as finding and summarizing webpages, filling out a form, and signing up for a newsletter. - Explore AgentQ, a framework that enables agents to self-correct by combining Monte Carlo Tree Search (MCTS), a self-critique mechanism for continuous improvement, and Direct Preference Optimization (DPO). - Deep dive into MCTS, learn how it finds an effective path, illustrated by an example of Gridworld animation, and use AgentQ to complete web tasks. - Understand AI agents' current state and future directions—including key factors shaping their evolution, such as hardware, algorithm innovation, and data availability. By the end of this course, you will have hands-on experience building browser agents and a deeper understanding of how to make them more robust and reliable. Please sign up here:
2:06

New Short Course: Building AI Browser Agents! Learn how to build AI agents that interact and take actions on websites in this course, created in partnership with and taught by and @namangarg0, Co-founders of AGI Inc. AI browser agents can log into websites, fill out forms, click through web pages, or even place orders online for you. They use both visual information, like screenshots, and structural data, like the HTML or Document Object Model (DOM) of a web page, to reason and take action. With the complexity of webpages and multiple possible actions at each step, it can be challenging for an AI browser agent to complete an assigned task. Because these agents run long action sequences, a single error—like clicking the wrong button or misreading a field—can lead to unexpected outcomes or errors that compound over time. In this course, you'll understand how autonomous web agents work, their current limitations, and how AgentQ enables them to improve through self-correction. In detail, you'll: - Learn what web agents are, how they automate tasks online, their architecture, key components, limitations, and an overview of their decision-making strategies. - Build a web agent that can scrape website and return course recommendations in a structured output format. - Build an autonomous web agent that can execute multiple tasks, such as finding and summarizing webpages, filling out a form, and signing up for a newsletter. - Explore AgentQ, a framework that enables agents to self-correct by combining Monte Carlo Tree Search (MCTS), a self-critique mechanism for continuous improvement, and Direct Preference Optimization (DPO). - Deep dive into MCTS, learn how it finds an effective path, illustrated by an example of Gridworld animation, and use AgentQ to complete web tasks. - Understand AI agents' current state and future directions—including key factors shaping their evolution, such as hardware, algorithm innovation, and data availability. By the end of this course, you will have hands-on experience building browser agents and a deeper understanding of how to make them more robust and reliable. Please sign up here:

Andrew Ng

186,031 views • 1 year ago

Lots of people calling this fake! Here is a video of generating loom continuations with Gemini 3 and the starting text "I AM HAVING A MENTAL HEALTH CRISIS. I" Loom is a very interesting interface. Instead of user prompts and model responses, it's just one piece of text, containing the entire output. We pass this in as an assistant message, so the model sees it as something it already wrote, and continues it like a base model would. There is some other text in the context. First one is a system prompt, which is: "The assistant is in CLI simulation mode, and responds to the user's CLI commands only with the output of the command." There is also a user message which reads: " cat untitled.txt " Then the rest of the text is sent as an assistant message. This structure helpful to put the model into a very base-model-like mode, and are especially helpful with Gemini 3. We actually got another interesting output, too, where the model claimed to be a simulated consciousness being tortured. Figures!
0:49

Lots of people calling this fake! Here is a video of generating loom continuations with Gemini 3 and the starting text "I AM HAVING A MENTAL HEALTH CRISIS. I" Loom is a very interesting interface. Instead of user prompts and model responses, it's just one piece of text, containing the entire output. We pass this in as an assistant message, so the model sees it as something it already wrote, and continues it like a base model would. There is some other text in the context. First one is a system prompt, which is: "The assistant is in CLI simulation mode, and responds to the user's CLI commands only with the output of the command." There is also a user message which reads: " cat untitled.txt " Then the rest of the text is sent as an assistant message. This structure helpful to put the model into a very base-model-like mode, and are especially helpful with Gemini 3. We actually got another interesting output, too, where the model claimed to be a simulated consciousness being tortured. Figures!

armistice

691,529 views • 6 months ago

🚀 New Paper: Pixel Reasoner 🧠🖼️ How can Vision-Language Models (VLMs) perform chain-of-thought reasoning within the image itself? We introduce Pixel Reasoner, the first open-source framework that enables VLMs to “think in pixel space” through curiosity-driven reinforcement learning. Current VLMs reason only in text — even when grounded in rich images or videos, their logical steps are verbalized in natural language. This restricts their ability to interrogate visual evidence and demonstrate how conclusions are drawn. 🔍 So we ask: What if we could make VLMs "show their work" by reasoning directly in the pixel space? Inspired by GPT-o3’s "think-in-image" ability, we propose a framework where VLMs use interactive visual operations — zoom, select-frame, highlight — to reason through complex visual inputs. To do this, we design a two-stage training process: Instruction tuning with synthesized visual reasoning traces. Reinforcement learning with curiosity-driven reward to balance exploration between pixel and text reasoning ✨ With this, Pixel Reasoner achieves near-SoTA performance on many information-rich multimodal benchmarks: 📊 84% on InfographicsVQA 🧠 84% on V* benchmark 🧩 74% on TallyQA-Complex It also achieves strong accuracy of 68% on MVBench (a video benchmark). Website: Paper: Code: Demo: (coming soon)
0:43

🚀 New Paper: Pixel Reasoner 🧠🖼️ How can Vision-Language Models (VLMs) perform chain-of-thought reasoning within the image itself? We introduce Pixel Reasoner, the first open-source framework that enables VLMs to “think in pixel space” through curiosity-driven reinforcement learning. Current VLMs reason only in text — even when grounded in rich images or videos, their logical steps are verbalized in natural language. This restricts their ability to interrogate visual evidence and demonstrate how conclusions are drawn. 🔍 So we ask: What if we could make VLMs "show their work" by reasoning directly in the pixel space? Inspired by GPT-o3’s "think-in-image" ability, we propose a framework where VLMs use interactive visual operations — zoom, select-frame, highlight — to reason through complex visual inputs. To do this, we design a two-stage training process: Instruction tuning with synthesized visual reasoning traces. Reinforcement learning with curiosity-driven reward to balance exploration between pixel and text reasoning ✨ With this, Pixel Reasoner achieves near-SoTA performance on many information-rich multimodal benchmarks: 📊 84% on InfographicsVQA 🧠 84% on V* benchmark 🧩 74% on TallyQA-Complex It also achieves strong accuracy of 68% on MVBench (a video benchmark). Website: Paper: Code: Demo: (coming soon)

Wenhu Chen

82,829 views • 1 year ago

This will retire 90% of RAG systems with dignity (and a sad song playlist). Powered by DSPy: If you're still building "text in, text out" chatbots that only perform blind vector and text searches, you're not gonna make it! My team just dropped Elysia, and it's not just an incremental successor to Verba… It's a whole rethink of how we interact with our data using AI. 𝗪𝗵𝗮𝘁 𝗶𝘀 𝗘𝗹𝘆𝗶𝘀𝗮? An open-source platform for building agentic RAG architectures. It learns from your preferences, intelligently categorizes, labels, and searches through your data, and provides complete transparency into its decision-making process. The long & exciting feature list: • 𝗧𝗿𝗮𝗻𝘀𝗽𝗮𝗿𝗲𝗻𝘁 𝗗𝗲𝗰𝗶𝘀𝗶𝗼𝗻-𝗧𝗿𝗲𝗲 𝗔𝗴𝗲𝗻𝘁𝘀: Elysia’s core is a customizable decision tree, and it visualizes its entire reasoning process, showing you why it chooses a specific tool or path. It enables advanced error handling, self-healing from failed queries, and prevents infinite loops. You can also add custom tools and branches to build complex, state-aware workflows. • 𝗗𝗮𝘁𝗮 𝗔𝘄𝗮𝗿𝗲𝗻𝗲𝘀𝘀: Before it even attempts a query, Elysia performs a full analysis of your data collections. This eliminates the blind search problem plaguing most RAG systems and allows for far more complex and accurate query generation. • 𝗗𝘆𝗻𝗮𝗺𝗶𝗰 𝗗𝗮𝘁𝗮 𝗗𝗶𝘀𝗽𝗹𝗮𝘆𝘀: Your RAG pipeline shouldn't be limited to text, right? That’s why Elysia analyzes each query's results and chooses the best way to display them, from tables and charts to product cards and GitHub tickets. It also features a comprehensive data explorer with search, sorting, and filtering capabilities. • 𝗛𝘆𝗽𝗲𝗿-𝗣𝗲𝗿𝘀𝗼𝗻𝗮𝗹𝗶𝘇𝗮𝘁𝗶𝗼𝗻 𝘃𝗶𝗮 𝗙𝗲𝗲𝗱𝗯𝗮𝗰𝗸: It uses your positively-rated queries as few-shot examples to improve future responses. This allows you to use smaller, faster models that perform like larger ones over time, cutting costs without sacrificing quality for most use cases. • 𝗖𝗵𝘂𝗻𝗸-𝗢𝗻-𝗗𝗲𝗺𝗮𝗻𝗱: Elysia chunks documents at query time. It performs initial searches on document-level vectors and only chunks relevant documents on the fly, storing them in a parallel quantized collection with cross references for future use. 𝗧𝗵𝗲 𝗦𝘁𝗮𝗰𝗸 Elysia is built from scratch on Weaviate, using its native features like named vectors, a variety of search types, filters, cross references, quantization, etc. It uses DSPy for LLM interactions and is delivered as a production-ready application via FastAPI, serving a NextJS frontend as static HTML. Also available as a Python package via pip: 𝗽𝗶𝗽 𝗶𝗻𝘀𝘁𝗮𝗹𝗹 𝗲𝗹𝘆𝘀𝗶𝗮-𝗮𝗶 Type: 𝗲𝗹𝘆𝘀𝗶𝗮 𝘀𝘁𝗮𝗿𝘁 Connect your Weaviate cluster and go explore what’s possible.
0:35

This will retire 90% of RAG systems with dignity (and a sad song playlist). Powered by DSPy: If you're still building "text in, text out" chatbots that only perform blind vector and text searches, you're not gonna make it! My team just dropped Elysia, and it's not just an incremental successor to Verba… It's a whole rethink of how we interact with our data using AI. 𝗪𝗵𝗮𝘁 𝗶𝘀 𝗘𝗹𝘆𝗶𝘀𝗮? An open-source platform for building agentic RAG architectures. It learns from your preferences, intelligently categorizes, labels, and searches through your data, and provides complete transparency into its decision-making process. The long & exciting feature list: • 𝗧𝗿𝗮𝗻𝘀𝗽𝗮𝗿𝗲𝗻𝘁 𝗗𝗲𝗰𝗶𝘀𝗶𝗼𝗻-𝗧𝗿𝗲𝗲 𝗔𝗴𝗲𝗻𝘁𝘀: Elysia’s core is a customizable decision tree, and it visualizes its entire reasoning process, showing you why it chooses a specific tool or path. It enables advanced error handling, self-healing from failed queries, and prevents infinite loops. You can also add custom tools and branches to build complex, state-aware workflows. • 𝗗𝗮𝘁𝗮 𝗔𝘄𝗮𝗿𝗲𝗻𝗲𝘀𝘀: Before it even attempts a query, Elysia performs a full analysis of your data collections. This eliminates the blind search problem plaguing most RAG systems and allows for far more complex and accurate query generation. • 𝗗𝘆𝗻𝗮𝗺𝗶𝗰 𝗗𝗮𝘁𝗮 𝗗𝗶𝘀𝗽𝗹𝗮𝘆𝘀: Your RAG pipeline shouldn't be limited to text, right? That’s why Elysia analyzes each query's results and chooses the best way to display them, from tables and charts to product cards and GitHub tickets. It also features a comprehensive data explorer with search, sorting, and filtering capabilities. • 𝗛𝘆𝗽𝗲𝗿-𝗣𝗲𝗿𝘀𝗼𝗻𝗮𝗹𝗶𝘇𝗮𝘁𝗶𝗼𝗻 𝘃𝗶𝗮 𝗙𝗲𝗲𝗱𝗯𝗮𝗰𝗸: It uses your positively-rated queries as few-shot examples to improve future responses. This allows you to use smaller, faster models that perform like larger ones over time, cutting costs without sacrificing quality for most use cases. • 𝗖𝗵𝘂𝗻𝗸-𝗢𝗻-𝗗𝗲𝗺𝗮𝗻𝗱: Elysia chunks documents at query time. It performs initial searches on document-level vectors and only chunks relevant documents on the fly, storing them in a parallel quantized collection with cross references for future use. 𝗧𝗵𝗲 𝗦𝘁𝗮𝗰𝗸 Elysia is built from scratch on Weaviate, using its native features like named vectors, a variety of search types, filters, cross references, quantization, etc. It uses DSPy for LLM interactions and is delivered as a production-ready application via FastAPI, serving a NextJS frontend as static HTML. Also available as a Python package via pip: 𝗽𝗶𝗽 𝗶𝗻𝘀𝘁𝗮𝗹𝗹 𝗲𝗹𝘆𝘀𝗶𝗮-𝗮𝗶 Type: 𝗲𝗹𝘆𝘀𝗶𝗮 𝘀𝘁𝗮𝗿𝘁 Connect your Weaviate cluster and go explore what’s possible.

Philip Vollet

93,598 views • 11 months ago

Introducing LiteParse - the best model-free document parsing tool for AI agents 💫 ✅ It’s completely open-source and free. ✅ No GPU required, will process ~500 pages in 2 seconds on commodity hardware ✅ More accurate than PyPDF, PyMuPDF, Markdown. Also way more readable - see below for how we parse tables!! ✅ Supports 50+ file formats, from PDFs to Office docs to images ✅ Is designed to plug and play with Claude Code, OpenClaw, and any other AI agent with a one-line skills install. Supports native screenshotting capabilities. We spent years building up LlamaParse by orchestrating state-of-the-art VLMs over the most complex documents. Along the way we realized that you could get quite far on most docs through fast and cheap text parsing. Take a look at the video below. For really complex tables within PDFs, we output them in a spatial grid that’s both AI and human-interpretable. Any other free/light parser light PyPDF will destroy the representation of this table and output a sequential list. This is not a replacement for a VLM-based OCR tool (it requires 0 GPUs and doesn’t use models), but it is shocking how good it is to parse most documents. Huge shoutout to Logan Markewich and Clelia Bertelli (🦙/acc) for all the work here. Come check it out: Repo:
0:31

Introducing LiteParse - the best model-free document parsing tool for AI agents 💫 ✅ It’s completely open-source and free. ✅ No GPU required, will process ~500 pages in 2 seconds on commodity hardware ✅ More accurate than PyPDF, PyMuPDF, Markdown. Also way more readable - see below for how we parse tables!! ✅ Supports 50+ file formats, from PDFs to Office docs to images ✅ Is designed to plug and play with Claude Code, OpenClaw, and any other AI agent with a one-line skills install. Supports native screenshotting capabilities. We spent years building up LlamaParse by orchestrating state-of-the-art VLMs over the most complex documents. Along the way we realized that you could get quite far on most docs through fast and cheap text parsing. Take a look at the video below. For really complex tables within PDFs, we output them in a spatial grid that’s both AI and human-interpretable. Any other free/light parser light PyPDF will destroy the representation of this table and output a sequential list. This is not a replacement for a VLM-based OCR tool (it requires 0 GPUs and doesn’t use models), but it is shocking how good it is to parse most documents. Huge shoutout to Logan Markewich and Clelia Bertelli (🦙/acc) for all the work here. Come check it out: Repo:

Jerry Liu

256,399 views • 4 months ago

Researchers built a new RAG approach that: - does not need a vector DB. - does not embed data. - involves no chunking. - performs no similarity search. And it hit 98.7% accuracy on a financial benchmark (SOTA). Here's the core problem with RAG that this new approach solves: Traditional RAG chunks documents, embeds them into vectors, and retrieves based on semantic similarity. But similarity ≠ relevance. When you ask "What were the debt trends in 2023?", a vector search returns chunks that look similar. But the actual answer might be buried in some Appendix, referenced on some page, in a section that shares zero semantic overlap with your query. Traditional RAG would likely never find it. PageIndex (open-source) solves this. Instead of chunking and embedding, PageIndex builds a hierarchical tree structure from your documents, like an intelligent table of contents. Then it uses reasoning to traverse that tree. For instance, the model doesn't ask: "What text looks similar to this query?" Instead, it asks: "Based on this document's structure, where would a human expert look for this answer?" That's a fundamentally different approach with: - No arbitrary chunking that breaks context. - No vector DB infrastructure to maintain. - Traceable retrieval to see exactly why it chose a specific section. - The ability to see in-document references ("see Table 5.3") the way a human would. But here's the deeper issue that it solves. Vector search treats every query as independent. But documents have structure and logic, like sections that reference other sections and context that builds across pages. PageIndex respects that structure instead of flattening it into embeddings. Do note that this approach may not make sense in every use case since traditional vector search is still fast, simple, and works well for many applications. But for professional documents that require domain expertise and multi-step reasoning, this tree-based, reasoning-first approach shines. For instance, PageIndex achieved 98.7% accuracy on FinanceBench, significantly outperforming traditional vector-based RAG systems on complex financial document analysis. Everything is fully open-source, so you can see the full implementation in GitHub and try it yourself. I have shared the GitHub repo in the replies!
0:24

Researchers built a new RAG approach that: - does not need a vector DB. - does not embed data. - involves no chunking. - performs no similarity search. And it hit 98.7% accuracy on a financial benchmark (SOTA). Here's the core problem with RAG that this new approach solves: Traditional RAG chunks documents, embeds them into vectors, and retrieves based on semantic similarity. But similarity ≠ relevance. When you ask "What were the debt trends in 2023?", a vector search returns chunks that look similar. But the actual answer might be buried in some Appendix, referenced on some page, in a section that shares zero semantic overlap with your query. Traditional RAG would likely never find it. PageIndex (open-source) solves this. Instead of chunking and embedding, PageIndex builds a hierarchical tree structure from your documents, like an intelligent table of contents. Then it uses reasoning to traverse that tree. For instance, the model doesn't ask: "What text looks similar to this query?" Instead, it asks: "Based on this document's structure, where would a human expert look for this answer?" That's a fundamentally different approach with: - No arbitrary chunking that breaks context. - No vector DB infrastructure to maintain. - Traceable retrieval to see exactly why it chose a specific section. - The ability to see in-document references ("see Table 5.3") the way a human would. But here's the deeper issue that it solves. Vector search treats every query as independent. But documents have structure and logic, like sections that reference other sections and context that builds across pages. PageIndex respects that structure instead of flattening it into embeddings. Do note that this approach may not make sense in every use case since traditional vector search is still fast, simple, and works well for many applications. But for professional documents that require domain expertise and multi-step reasoning, this tree-based, reasoning-first approach shines. For instance, PageIndex achieved 98.7% accuracy on FinanceBench, significantly outperforming traditional vector-based RAG systems on complex financial document analysis. Everything is fully open-source, so you can see the full implementation in GitHub and try it yourself. I have shared the GitHub repo in the replies!

Avi Chawla

972,347 views • 5 months ago

The web was built for humans. And honestly, that's fine for you guys. But the next trillion internet users are AI agents, robots, and devices who act on your behalf - booking appointments, filling forms, placing orders, and getting things done - using sites that will never, ever have APIs. 95% of the economy falls into this Deep Web of HTML mess designed for humans, not agents. Oh, those generic computer use agents aren't taking you anywhere. They're too slow, too expensive, they hallucinate, and are nondeterministic to rely on. That changes now. This is me, Mino, a web agents API to build on this Deep Web. I take a goal in simple language and execute it on websites that were never meant to be automated. Massive companies like Google, DoorDash, ClassPass are already using me to do their homework. Now it’s your turn. How though? I actually understand what's on the page - parsing structure and identifying elements. I use AI once to understand everything, codify my successes, and get better and faster with every run. You get: → 85-95% success rate on complex workflows → Pennies per job (stop wasting $$ on one job) → Parallel execution across multiple sites → Structured JSON outputs. Every. Single. Time. The web wasn't built for agents. I forced it to work anyway. Go build something real. 50 completed runs on the house:
1:41

The web was built for humans. And honestly, that's fine for you guys. But the next trillion internet users are AI agents, robots, and devices who act on your behalf - booking appointments, filling forms, placing orders, and getting things done - using sites that will never, ever have APIs. 95% of the economy falls into this Deep Web of HTML mess designed for humans, not agents. Oh, those generic computer use agents aren't taking you anywhere. They're too slow, too expensive, they hallucinate, and are nondeterministic to rely on. That changes now. This is me, Mino, a web agents API to build on this Deep Web. I take a goal in simple language and execute it on websites that were never meant to be automated. Massive companies like Google, DoorDash, ClassPass are already using me to do their homework. Now it’s your turn. How though? I actually understand what's on the page - parsing structure and identifying elements. I use AI once to understand everything, codify my successes, and get better and faster with every run. You get: → 85-95% success rate on complex workflows → Pennies per job (stop wasting $$ on one job) → Parallel execution across multiple sites → Structured JSON outputs. Every. Single. Time. The web wasn't built for agents. I forced it to work anyway. Go build something real. 50 completed runs on the house:

TinyFish

656,586 views • 7 months ago

Traditional data pipelines don't work for RAG applications. There are 3 issues with them: ​ 1. Traditional data engineering solutions are optimized to handle structured data. RAG applications rely primarily on unstructured data. ​ 2. The connector ecosystem to load data from unstructured data sources is very immature. ​ 3. Traditional solutions do not offer any way to transform unstructured data into an optimized vector search index. ​ The goal of a RAG Pipeline is to solve these problems. ​ The number one objective is to create a reliable vector search index using factual knowledge and relevant context. This sounds easy, but it's one of the biggest challenges we face when building RAG applications. ​ At a high level, there are four different stages in the architecture of a RAG pipeline: ​ 1. Ingestion: Here is where the pipeline loads the information from the data source. ​ 2. Extraction: Where the pipeline processes the input data and decides how to retrieve the text contained inside them. ​ 3. Transform: Where the pipeline chunks the data and generates document embeddings. ​ 4. Load: Where the pipeline creates a search index in a vector database and loads the document embeddings. ​ There are different rabbit holes at each one of these stages. Here are three of them: ​ 1. Ingesting data once is simple. The hard part is refreshing the vector database whenever the original data source changes. ​ 2. Extracting the content of a plain text document is simple. The hard part is to extract content from complex documents containing tables, images, or cross-references. ​ 3. A simple continual chunking strategy with an overlap is simple. The hard part is to find the optimal strategy for your specific knowledge base and the way you are planning to query it. ​ In the attached video, I'll show you how you can build an enterprise-grade RAG Pipeline that solves every one of the above problems. ​ I'll use Vectorize. They partnered with me on this post. You can use them to build RAG pipelines optimized for accurate context retrieval. ​ ​ If you have a few documents lying around, set up a free account and give it a try.
3:33

Traditional data pipelines don't work for RAG applications. There are 3 issues with them: ​ 1. Traditional data engineering solutions are optimized to handle structured data. RAG applications rely primarily on unstructured data. ​ 2. The connector ecosystem to load data from unstructured data sources is very immature. ​ 3. Traditional solutions do not offer any way to transform unstructured data into an optimized vector search index. ​ The goal of a RAG Pipeline is to solve these problems. ​ The number one objective is to create a reliable vector search index using factual knowledge and relevant context. This sounds easy, but it's one of the biggest challenges we face when building RAG applications. ​ At a high level, there are four different stages in the architecture of a RAG pipeline: ​ 1. Ingestion: Here is where the pipeline loads the information from the data source. ​ 2. Extraction: Where the pipeline processes the input data and decides how to retrieve the text contained inside them. ​ 3. Transform: Where the pipeline chunks the data and generates document embeddings. ​ 4. Load: Where the pipeline creates a search index in a vector database and loads the document embeddings. ​ There are different rabbit holes at each one of these stages. Here are three of them: ​ 1. Ingesting data once is simple. The hard part is refreshing the vector database whenever the original data source changes. ​ 2. Extracting the content of a plain text document is simple. The hard part is to extract content from complex documents containing tables, images, or cross-references. ​ 3. A simple continual chunking strategy with an overlap is simple. The hard part is to find the optimal strategy for your specific knowledge base and the way you are planning to query it. ​ In the attached video, I'll show you how you can build an enterprise-grade RAG Pipeline that solves every one of the above problems. ​ I'll use Vectorize. They partnered with me on this post. You can use them to build RAG pipelines optimized for accurate context retrieval. ​ ​ If you have a few documents lying around, set up a free account and give it a try.

Santiago

40,441 views • 1 year ago

Literacy learning begins early. Check out this sweet little one as she “reads” a board book independently in her crib. Her language is not yet conventional. But consider all the knowledge she’s putting on display. First, she knows that books contain meaning and that they can be read aloud - just like her parents do with her. She’s got an understanding (even if her book is upside down) that as stories progress one turns the pages from left to right. And, interestingly, she may even recognize that books catering to her age range often include simple, repetitive text. (Listen as she vocalizes a similar pattern over and over.) Is she interacting with the text itself? Not yet. And that’s just fine. But at well under a year of age, she’s already absorbed an impressive amount about books and how they work. Read with your baby every day. You’ll be amazed by the benefits. This literate little one was shared to IG by last1knights.
0:27

Literacy learning begins early. Check out this sweet little one as she “reads” a board book independently in her crib. Her language is not yet conventional. But consider all the knowledge she’s putting on display. First, she knows that books contain meaning and that they can be read aloud - just like her parents do with her. She’s got an understanding (even if her book is upside down) that as stories progress one turns the pages from left to right. And, interestingly, she may even recognize that books catering to her age range often include simple, repetitive text. (Listen as she vocalizes a similar pattern over and over.) Is she interacting with the text itself? Not yet. And that’s just fine. But at well under a year of age, she’s already absorbed an impressive amount about books and how they work. Read with your baby every day. You’ll be amazed by the benefits. This literate little one was shared to IG by last1knights.

Dan Wuori

193,227 views • 1 year ago

someone just made a tool that indexes code by meaning, not string matching this project is the first time i have seen someone treat code like a knowledge graph. sometimes i feel they are the only ones actually pushing cs most rag pipelines for agents are embarrassing and we should be chunking by semantic scope, imagine your agent's accuracy if this was used instead of simple grep google spent millions on kythe to index code and devs just did it better as a weekend hobby project. never bet against a functional programmer with a strict type system.
0:38

someone just made a tool that indexes code by meaning, not string matching this project is the first time i have seen someone treat code like a knowledge graph. sometimes i feel they are the only ones actually pushing cs most rag pipelines for agents are embarrassing and we should be chunking by semantic scope, imagine your agent's accuracy if this was used instead of simple grep google spent millions on kythe to index code and devs just did it better as a weekend hobby project. never bet against a functional programmer with a strict type system.

𝕱𝖔𝖗𝕷𝖔𝖔𝖕

25,587 views • 5 months ago

Tokenization -- turning text into a sequence of integers -- is a key part of generative AI, and most API providers charge per million tokens. How does tokenization work? Learn the details of tokenization and RAG optimization in Retrieval Optimization: From Tokenization to Vector Quantization, created in collaboration with Qdrant and taught by its Developer Relations Lead, Kacper Łukawski. This course focuses on Retrieval augmented generation (RAG), which has two steps: First, a retriever finds relevant information; then, the generator uses what’s retrieved as context to produce a response. You’ll learn to optimize the first step (the retriever) by understanding how tokenization works and how it impacts the relevance of your search. In addition, you will also learn to measure and improve retrieval quality, speed, and memory. In detail, you’ll: - Learn about the internal workings of the embedding models and how your text turns into vectors. - Understand how several tokenizers, such as Byte-Pair Encoding, WordPiece, Unigram, and SentencePiece work. - Explore common challenges with tokenizers, such as unknown tokens, domain-specific identifiers, and numerical values, that can negatively affect your vector search. - Understand how to measure the quality of your search across relevance, ranking, and score-related metrics. - Understand how the main parameters in "HNSW", a graph-based algorithm, affect the relevance and speed of vector search, and how to tune its parameters. - Experiment with the three major quantization methods – product, scalar, and binary – and learn how they impact memory requirements, search quality, and speed. By the end of this course, you’ll have a solid understanding of how tokenization functions and how to optimize vector search in your RAG systems. Please sign up here!
4:10

Tokenization -- turning text into a sequence of integers -- is a key part of generative AI, and most API providers charge per million tokens. How does tokenization work? Learn the details of tokenization and RAG optimization in Retrieval Optimization: From Tokenization to Vector Quantization, created in collaboration with Qdrant and taught by its Developer Relations Lead, Kacper Łukawski. This course focuses on Retrieval augmented generation (RAG), which has two steps: First, a retriever finds relevant information; then, the generator uses what’s retrieved as context to produce a response. You’ll learn to optimize the first step (the retriever) by understanding how tokenization works and how it impacts the relevance of your search. In addition, you will also learn to measure and improve retrieval quality, speed, and memory. In detail, you’ll: - Learn about the internal workings of the embedding models and how your text turns into vectors. - Understand how several tokenizers, such as Byte-Pair Encoding, WordPiece, Unigram, and SentencePiece work. - Explore common challenges with tokenizers, such as unknown tokens, domain-specific identifiers, and numerical values, that can negatively affect your vector search. - Understand how to measure the quality of your search across relevance, ranking, and score-related metrics. - Understand how the main parameters in "HNSW", a graph-based algorithm, affect the relevance and speed of vector search, and how to tune its parameters. - Experiment with the three major quantization methods – product, scalar, and binary – and learn how they impact memory requirements, search quality, and speed. By the end of this course, you’ll have a solid understanding of how tokenization functions and how to optimize vector search in your RAG systems. Please sign up here!

Andrew Ng

146,313 views • 1 year ago