{ "source_file": "0930", "source_platform": "Medium", "author": "Julian Yip", "title_listed": "Bringing Vision-Language Intelligence to RAG with ColPali", "keywords": "", "domain": "大模型算法", "content_type": "深度技术解析", "is_paid_flag": "否", "url": "https://towardsdatascience.com/bringing-vision-language-intelligence-to-rag-with-colpali/", "publish_date": "2025-10-29", "corpus_stage": "", "notes": "", "id": "191976f37240", "collected_at": "2026-10-02T06:39:18", "title": "Bringing Vision-Language Intelligence to RAG with ColPali | Towards Data Science", "body_markdown": "![Revisit consent button](../images/body-df29beacd7496220cc94.svg)\nWe value your privacy\nWe use cookies to enhance your browsing experience, serve personalised ads or content, and analyse our traffic. By clicking \"Accept All\", you consent to our use of cookies.\nCustomise Reject All Accept All\nCustomise Consent Preferences ![](../images/body-2159bb67f3de8ad870eb.svg)\nWe use cookies to help you navigate efficiently and perform certain functions. You will find detailed information about all cookies under each consent category below.\nThe cookies that are categorised as \"Necessary\" are stored on your browser as they are essential for enabling the basic functionalities of the site. ... Show more\nNecessaryAlways Active\nNecessary cookies are required to enable the basic features of this site, such as providing secure log-in or adjusting your consent preferences. These cookies do not store any personally identifiable data.\n * Cookie\nBCTempID\n * Duration\n10 minutes\n * Description\nNo description available.\n\n\n * Cookie\n__cf_bm\n * Duration\n1 hour\n * Description\nThis cookie, set by Cloudflare, is used to support Cloudflare Bot Management. \n\n\n * Cookie\nAWSALBCORS\n * Duration\n7 days\n * Description\nAmazon Web Services set this cookie for load balancing.\n\n\n * Cookie\n_cfuvid\n * Duration\nsession\n * Description\nCloudflare sets this cookie to track users across sessions to optimize user experience by maintaining session consistency and providing personalized services\n\n\n * Cookie\nli_gc\n * Duration\n6 months\n * Description\nLinkedin set this cookie for storing visitor's consent regarding using cookies for non-essential purposes.\n\n\n * Cookie\n__hssrc\n * Duration\nsession\n * Description\nThis cookie is set by Hubspot whenever it changes the session cookie. The __hssrc cookie set to 1 indicates that the user has restarted the browser, and if the cookie does not exist, it is assumed to be a new session.\n\n\n * Cookie\n__hssc\n * Duration\n1 hour\n * Description\nHubSpot sets this cookie to keep track of sessions and to determine if HubSpot should increment the session number and timestamps in the __hstc cookie.\n\n\n * Cookie\nwpEmojiSettingsSupports\n * Duration\nsession\n * Description\nWordPress sets this cookie when a user interacts with emojis on a WordPress site. It helps determine if the user's browser can display emojis properly.\n\n\n * Cookie\nBCSessionID\n * Duration\n1 year 1 month 4 days\n * Description\nBlueconic sets this cookie as a unique identifier for the BlueConic profile.\n\n\n * Cookie\n_octo\n * Duration\n1 year\n * Description\nNo description available.\n\n\n * Cookie\nlogged_in\n * Duration\n1 year\n * Description\nNo description available.\n\n\n * Cookie\n__Secure-YEC\n * Duration\npast\n * Description\nYouTube sets this cookie to stores the user's video player preferences using embedded YouTube video\n\n\n * Cookie\n__eoi\n * Duration\n6 months\n * Description\nDescription is currently not available.\n\n\n * Cookie\nAWSALBTGCORS\n * Duration\n7 days\n * Description\nNo description available.\n\n\n * Cookie\nlogin-status-p\n * Duration\npast\n * Description\nDescription is currently not available.\n\n\n * Cookie\nAWSALBTG\n * Duration\n7 days\n * Description\nNo description available.\n\n\n * Cookie\ncsrf_token\n * Duration\nsession\n * Description\nNo description available.\n\n\n * Cookie\ntoken_v2\n * Duration\n1 day\n * Description\nDescription is currently not available.\n\n\n * Cookie\nD\n * Duration\n1 year\n * Description\nDescription is currently not available.\n\n\n * Cookie\nPHPSESSID\n * Duration\nsession\n * Description\nThis cookie is native to PHP applications. The cookie stores and identifies a user's unique session ID to manage user sessions on the website. The cookie is a session cookie and will be deleted when all the browser windows are closed.\n\n\n * Cookie\nVISITOR_PRIVACY_METADATA\n * Duration\n6 months\n * Description\nYouTube sets this cookie to store the user's cookie consent state for the current domain. \n\n\n * Cookie\ncookietest\n * Duration\nsession\n * Description\nThe cookietest cookie is typically used to determine whether the user's browser accepts cookies, essential for website functionality and user experience.\n\n\n * Cookie\n__Host-airtable-session\n * Duration\n1 year\n * Description\nThis cookie is used to enable us to integrate the services of Airtable.\n\n\n * Cookie\n__Host-airtable-session.sig\n * Duration\n1 year\n * Description\nThis cookie is used to enable us to integrate the services of Airtable.\n\n\n * Cookie\nm\n * Duration\n1 year 1 month 4 days\n * Description\nStripe sets this cookie for fraud prevention purposes. It identifies the device used to access the website, allowing the website to be formatted accordingly.\n\n\n * Cookie\nBIGipServer*\n * Duration\nsession\n * Description\nMarketo sets this cookie to collect information about the user's online activity and build a profile about their interests to provide advertisements relevant to the user. \n\n\n * Cookie\n__cfruid\n * Duration\nsession\n * Description\nCloudflare sets this cookie to identify trusted web traffic.\n\n\n * Cookie\n_GRECAPTCHA\n * Duration\n6 months\n * Description\nGoogle Recaptcha service sets this cookie to identify bots to protect the website against malicious spam attacks.\n\n\n * Cookie\n__Secure-YNID\n * Duration\n6 months\n * Description\nGoogle cookie used to protect user security and prevent fraud, especially during the login process.\n\n\n * Cookie\ncookieyes-consent\n * Duration\n1 year\n * Description\nCookieYes sets this cookie to remember users' consent preferences so that their preferences are respected on subsequent visits to this site. It does not collect or store any personal information about the site visitors.\n\n\nFunctional\nFunctional cookies help perform certain functionalities like sharing the content of the website on social media platforms, collecting feedback, and other third-party features.\n * Cookie\nlidc\n * Duration\n1 day\n * Description\nLinkedIn sets the lidc cookie to facilitate data center selection.\n\n\n * Cookie\nbrw\n * Duration\n1 year\n * Description\nNo description available.\n\n\n * Cookie\nbrwConsent\n * Duration\n5 minutes\n * Description\nDescription is currently not available.\n\n\n * Cookie\nWMF-Uniq\n * Duration\n1 year\n * Description\nDescription is currently not available.\n\n\n * Cookie\nloom_anon_comment\n * Duration\n1 year\n * Description\nNo description available.\n\n\n * Cookie\nloom_referral_video\n * Duration\nsession\n * Description\nDescription is currently not available.\n\n\n * Cookie\nVISITOR_INFO1_LIVE\n * Duration\n6 months\n * Description\nA cookie set by YouTube to measure bandwidth that determines whether the user gets the new or old player interface.\n\n\n * Cookie\nyt-remote-connected-devices\n * Duration\nNever Expires\n * Description\nYouTube sets this cookie to store the user's video preferences using embedded YouTube videos.\n\n\n * Cookie\nytidb::LAST_RESULT_ENTRY_KEY\n * Duration\nNever Expires\n * Description\nThe cookie ytidb::LAST_RESULT_ENTRY_KEY is used by YouTube to store the last search result entry that was clicked by the user. This information is used to improve the user experience by providing more relevant search results in the future.\n\n\n * Cookie\nyt-remote-device-id\n * Duration\nNever Expires\n * Description\nYouTube sets this cookie to store the user's video preferences using embedded YouTube videos.\n\n\n * Cookie\nyt-remote-session-name\n * Duration\nsession\n * Description\nThe yt-remote-session-name cookie is used by YouTube to store the user's video player preferences using embedded YouTube video.\n\n\n * Cookie\nyt-remote-fast-check-period\n * Duration\nsession\n * Description\nThe yt-remote-fast-check-period cookie is used by YouTube to store the user's video player preferences for embedded YouTube videos.\n\n\n * Cookie\nyt-remote-session-app\n * Duration\nsession\n * Description\nThe yt-remote-session-app cookie is used by YouTube to store user preferences and information about the interface of the embedded YouTube video player.\n\n\n * Cookie\nyt-remote-cast-available\n * Duration\nsession\n * Description\nThe yt-remote-cast-available cookie is used to store the user's preferences regarding whether casting is available on their YouTube video player.\n\n\n * Cookie\nyt-remote-cast-installed\n * Duration\nsession\n * Description\nThe yt-remote-cast-installed cookie is used to store the user's video player preferences using embedded YouTube video.\n\n\n * Cookie\ncp_session\n * Duration\n3 months\n * Description\nCodepen sets this cookie for Help systems found in the website.\n\n\n * Cookie\nloid\n * Duration\n1 year 1 month 4 days\n * Description\nThis cookie is set by the Reddit. The cookie enables the sharing of content from the website onto the social media platform.\n\n\nAnalytics\nAnalytical cookies are used to understand how visitors interact with the website. These cookies help provide information on metrics such as the number of visitors, bounce rate, traffic source, etc.\n * Cookie\n__hstc\n * Duration\n6 months\n * Description\nHubspot set this main cookie for tracking visitors. It contains the domain, initial timestamp (first visit), last timestamp (last visit), current timestamp (this visit), and session number (increments for each subsequent session).\n\n\n * Cookie\nhubspotutk\n * Duration\n6 months\n * Description\nHubSpot sets this cookie to keep track of the visitors to the website. This cookie is passed to HubSpot on form submission and used when deduplicating contacts.\n\n\n * Cookie\n_ga\n * Duration\n1 year 1 month 4 days\n * Description\nGoogle Analytics sets this cookie to calculate visitor, session and campaign data and track site usage for the site's analytics report. The cookie stores information anonymously and assigns a randomly generated number to recognise unique visitors.\n\n\n * Cookie\n_ga_*\n * Duration\n1 year 1 month 4 days\n * Description\nGoogle Analytics sets this cookie to store and count page views.\n\n\n * Cookie\n__Host-psifi.analyticsTrace\n * Duration\n6 hours\n * Description\nDescription is currently not available.\n\n\n * Cookie\n__Host-psifi.analyticsTraceV2\n * Duration\n6 hours\n * Description\nDescription is currently not available.\n\n\n * Cookie\n_gh_sess\n * Duration\nsession\n * Description\nGitHub sets this cookie for temporary application and framework state between pages like what step the user is on in a multiple step form.\n\n\n * Cookie\nYSC\n * Duration\nsession\n * Description\nYSC cookie is set by Youtube and is used to track the views of embedded videos on Youtube pages.\n\n\n * Cookie\najs_anonymous_id\n * Duration\n1 year\n * Description\nThis cookie is set by Segment to count the number of people who visit a certain site by tracking if they have visited before.\n\n\n * Cookie\nvuid\n * Duration\n1 year 1 month 4 days\n * Description\nVimeo installs this cookie to collect tracking information by setting a unique ID to embed videos on the website. \n\n\nPerformance\nPerformance cookies are used to understand and analyse the key performance indexes of the website which helps in delivering a better user experience for the visitors.\n * Cookie\nAWSALB\n * Duration\n7 days\n * Description\nAWSALB is an application load balancer cookie set by Amazon Web Services to map the session to the target.\n\n\n * Cookie\nacq\n * Duration\npast\n * Description\nDescription is currently not available.\n\n\n * Cookie\nacq.sig\n * Duration\npast\n * Description\nDescription is currently not available.\n\n\n * Cookie\nptc\n * Duration\n2 years\n * Description\nNo description available.\n\n\nAdvertisement\nAdvertisement cookies are used to provide visitors with customised advertisements based on the pages you visited previously and to analyse the effectiveness of the ad campaigns.\n * Cookie\nmuc_ads\n * Duration\n1 year 1 month 4 days\n * Description\nTwitter sets this cookie to collect user behaviour and interaction data to optimize the website.\n\n\n * Cookie\nguest_id_marketing\n * Duration\n1 year 1 month 4 days\n * Description\nTwitter sets this cookie to identify and track the website visitor.\n\n\n * Cookie\nguest_id_ads\n * Duration\n1 year 1 month 4 days\n * Description\nTwitter sets this cookie to identify and track the website visitor.\n\n\n * Cookie\npersonalization_id\n * Duration\n1 year 1 month 4 days\n * Description\nTwitter sets this cookie to integrate and share features for social media and also store information about how the user uses the website, for tracking and targeting.\n\n\n * Cookie\nguest_id\n * Duration\n1 year 1 month 4 days\n * Description\nTwitter sets this cookie to identify and track the website visitor. It registers if a user is signed in to the Twitter platform and collects information about ad preferences.\n\n\n * Cookie\nbcookie\n * Duration\n1 year\n * Description\nLinkedIn sets this cookie from LinkedIn share buttons and ad tags to recognize browser IDs.\n\n\n * Cookie\n__Secure-ROLLOUT_TOKEN\n * Duration\n6 months\n * Description\nYouTube sets this cookie to manage feature rollout and experimentation. It helps Google control which new features or interface changes are shown to users as part of testing and staged rollouts, ensuring consistent experience for a given user during an experiment.\n\n\n * Cookie\nyt.innertube::nextId\n * Duration\nNever Expires\n * Description\nYouTube sets this cookie to register a unique ID to store data on what videos from YouTube the user has seen.\n\n\n * Cookie\nyt.innertube::requests\n * Duration\nNever Expires\n * Description\nYouTube sets this cookie to register a unique ID to store data on what videos from YouTube the user has seen.\n\n\n * Cookie\nsession_tracker\n * Duration\nsession\n * Description\nThis cookie is set by the Reddit. This cookie is used to identify trusted web traffic. It also helps in adverstising on the website.\n\n\n * Cookie\nedgebucket\n * Duration\nsession\n * Description\nReddit sets this cookie to save the information about a log-on Reddit user, for the purpose of advertisement recommendations and updating the content.\n\n\n * Cookie\ndid\n * Duration\n1 year\n * Description\nArbor sets this cookie to show targeted ads to site visitors.This cookie expires after 2 months or 1 year.\n\n\nUncategorised\nOther uncategorised cookies are those that are being analysed and have not been classified into a category as yet.\nNo cookies to display.\nReject All Save My Preferences Accept All\nToggle Search[](https://towardsdatascience.com/)Toggle Mobile Navigation\n[](https://towardsdatascience.com/)\nPublish AI, ML & data-science insights to a global community of data professionals.\nSign in[Submit an Article](https://write.insightmediagroup.io)\n[Latest](https://towardsdatascience.com/latest/)[Editor's Picks](https://towardsdatascience.com/tag/editors-pick/)[Newsletter](https://towardsdatascience.com/tag/the-variable/)[ShipAI](https://towardsdatascience.com/shipai/)\n* * *\n[Write For TDS](https://towardsdatascience.com/submissions/)\n[LinkedIn](https://www.linkedin.com/company/towards-data-science/)[X](https://x.com/TDataScience)Toggle Search\n[Large Language Models](https://towardsdatascience.com/category/artificial-intelligence/large-language-models/)\n# Bringing Vision-Language Intelligence to RAG with ColPali\nUnlocking the value of non-textual contents in your knowledge base\n[Julian Yip](https://towardsdatascience.com/author/jyipkl/)\nOctober 29, 20257 min read\n![](../images/000.png)\nSource: Image by the author.\nIf you’ve ever tried building a RAG (Retrieval-Augmented Generation) application, you’re likely familiar with the challenges posed by tables and images. This article explores how to tackle these formats using Vision Language Models, specifically with the ColPali model.\nBut first, what exactly is RAG — and why do tables and images make it so difficult?\n## RAG and parsing\nImagine you’re faced with a question like:\n> What’s the our company’s policy for handling refund?\nA foundational LLM (Large Language Model) probably won’t be able to answer this, as such information is company-specific and typically not included in the model’s training data.\nLearn this step by step with the interactive [AI Engineer](https://roadmap.sh/ai-engineer) roadmap.\nThat’s why a common approach is to connect the LLM to a knowledge base — such as a SharePoint folder containing various internal documents. This allows the model to retrieve and incorporate relevant context, enabling it to answer questions that require specialized knowledge. This technique is known as Retrieval-Augmented Generation (RAG), and it often involves working with documents like PDFs.\nHowever, extracting the right information from a large and diverse knowledge base requires extensive document preprocessing. Common steps include:\n 1. **Parsing** : Parsing documents into texts and images, often assisted with Optical Character Recognition (OCR) tools like Tesseract. Tables are most often converted into texts\n 2. **Structure Preservation** : Maintain the structure of the document, including headings, paragraphs, by converting the extracted text into a format that retains context, such as Markdown\n 3. **Chunking** : Splitting or merging text passages, so that the contexts can be fed into the context window without causing the passages come across as disjointed\n 4. **Enriching** : Provide extra metadata e.g. extract keyword or provide summary to the chunks to ease discovery. Optionally, to also caption images with descriptive texts via multimodal LLM to make images searchable\n 5. **Embedding** : Embed the texts (and potentially the images too with multimodal embedding), and store them into a vector DB\n\n\nAs you can imagine, the process is highly complicated, involves a lot of experimentation, and is very brittle. Worse yet, even if we tried to do it as best as we could, this parsing might not actually work after all.\n## Why parsing often falls short\nTables and image often exist in PDFs. The below image shows how they are typically parsed for LLM’s consumption:\n![](../images/001.png)\nSource: Image by the author.\n * **Texts** are chunked\n * **Tables** are turned into texts, whatever contained within are copied without preserving table boundaries\n * **Images** are fed into multimodal LLM for text summary generation, or alternatively, the original image is fed into multimodal embedding model without needing to generate a text summary\n\n\nHowever, there are two inherent issues with such traditional approach.\n**#1. Complex tables cannot be simply be interpreted as texts** \nTaking this table as an example, we as human would interpret that a temperature change of **> 2˚C to 2.5˚C’**s implication on **Health** is **A rise of 2.3˚C by 2080 puts up to 270 million at risk from malaria**\n![](../images/002.png)\nSource: [The Impacts and Costs of Climate Change](https://climate.ec.europa.eu/system/files/2016-11/final_report2_en.pdf)\nHowever, if we turn this table into a text, it would look like this: `Temperature change Within EC target <(2˚C) >2˚C to 2.5˚C >3C Health Globally it is estimated that A rise of 2.3oC by 2080 puts A rise of 3.3oC by 2080 an average temperature rise up to 270 million at risk from would put up to 330...`\nThe result is a jumbled block of text with no discernible meaning. Even for a human reader, it’s impossible to extract any meaningful insight from it. When this kind of text is fed into a Large Language Model (LLM), it also fails to produce an accurate interpretation.\n**#2. Disassociation between texts and images** \nThe description of the image is often included in texts and they are inseparable from one another. Taking the below as an example, we know the chart represents the “Modelled Costs of Climate Change with Different Pure Rate of Time Preference and declining discount rate schemes (no equity weighting)”\n![](../images/body-8ae18c4f8013be994cdc.png)\nSource: [The Impacts and Costs of Climate Change](https://climate.ec.europa.eu/system/files/2016-11/final_report2_en.pdf)\nHowever, as this is parsed, the image description (parsed text) will be disassociated with the image (parsed chart). So we can expect, during RAG, the image would not be retrieved as input when we raise a question like “what’s the cost of climate change?”\n![](../images/body-49aa973d5b1af5fce0e3.png)\nSo, even if we attempt to engineer solutions that preserve as much information as possible during parsing, they often fall short when faced with real-world scenarios.\nGiven how critical parsing is in RAG applications, does this mean RAG agents are destined to fail when working with complex documents? Absolutely not. With ColPali, we have a more refined and effective approach to handling them.\n## What’s ColPali?\nThe core premise of ColPali is simple: Human read PDF as pages, not “chunks”, so it makes sense to treat PDF as such: Instead of going through the messy process of parsing, we just turn the PDF pages into images, and use that as context for LLM to provide an answer.\n![](../images/body-f5325dd8bc574d543594.png)\nNow, the idea of embedding images using multimodal models isn’t new — it’s a common technique. So what makes ColPali stand out? The key lies in its inspiration from ColBERT, a model that embeds inputs into multi-vectors, enabling more precise and efficient search.\nBefore diving into ColPali’s capabilities, let me briefly digress to explain what ColBERT is all about.\n#### ColBERT: Granular, context-aware embedding for texts\nColBERT is a text embedding and reranking technique that leverage on multi-vectors to enhance search accuracy for texts.\nLet’s consider this case: we have this question: `is Paul vegan?`, we need to identify which text chuck contains the relevant information.\n![](../images/body-1bf946b8ea8d564ad7af.png)\n_Highlighted in yellow are texts which contain information about Paul_\nIdeally, we should identify `Text Chunk A` as the most relevant one. But if we use a single-vector embedding model (text-ada-002), it will return `Text Chunk B` instead.\nThe reason lies in how single-vector bi-encoders — like text-ada-002 — operate. They attempt to compress an entire sentence into a single vector, without encoding individual words in a context-aware manner. In contrast, ColBERT **embeds each word with contextual awareness** , resulting in a richer, **multi-vector** representation that captures more nuanced information.\n![](../images/body-ca9a1cb1493bb7b72cc7.png)\nNumbers in the vectors are illustrative and do not represents the actual values\n#### ColPali: ColBERT’s brother for handling document-like images\nColPali follows a similar philosophy but applies it to document-like images. Just as ColBERT breaks down text and embeds each word individually, ColPali divides an image into patches and generates embeddings for each patch. This approach preserves more of the **image’s contextual detail, enabling more accurate and meaningful interpretation**.\n![](../images/body-4568a9ceaeef3700754d.png)\nApart from higher retrieval accuracy, benefits of ColPali includes:\n 1. **Explainability** : ColPali enables word-level comparison between the query and individual image patches of a document. This allows us to clearly understand and justify why a particular document is deemed more relevant.\n 2. **Reduced Development Effort & Greater Robustness**: By eliminating the need for complex preprocessing pipelines — such as chunking, OCR, and layout parsing — ColPali significantly reduces development time and minimizes potential points of failure.\n 3. **Performance Gains** : Embedding and retrieval processes are faster, resulting in improved overall system responsiveness.\n\n\nNow you know what ColPali is, let’s dive into the code and see if ColPali can resolve the challenges we mentioned earlier!\n## Illustration\nMy code can be found in my [Github](https://github.com/yip-kl?tab=repositories). A few words about it:\n * **Instance** : Running the code requires a machine with A100 GPU\n * **Embedding model** : ColPali has a number of variants, I am using `vidore/colqwen2-v0.1` for demonstration. You can refer to the leaderboard [here](https://huggingface.co/spaces/vidore/vidore-leaderboard) and switch to other models\n * **Agent:** \n**- LLM** : I am using OpenRouter to consume LLM, and the agent is powered by `GPT-4o`. You can substitute it with any multimodal LLM that can take image as input \n- **Orchestration** : `LangGraph` is used to develop the agent\n * **Library used for converting PDF into images** : `pdf2image` which is a wrapper for `poppler`. So please make sure you also have `poppler` installed\n * **Sample data** : “The Impacts and Costs of Climate Change” written by Paul Watkiss et al, which is publicly available [here](https://climate.ec.europa.eu/system/files/2016-11/final_report2_en.pdf)\n\n\n### Results\nThe code is pretty straightforward, so I will jump into conclusion instead: How well can ColPali deal with the problems we have seen in the “Why parsing often falls short?” section?\n**#1. Complex tables cannot be simply be interpreted as texts**\n**Question** : What’s the health impact if temperature rises above pre-industrial level by 3 degrees?\n**Answer** : If the temperature rises above pre-industrial levels by 3 degrees Celsius, the health impact would be significant. `Specifically, it is estimated that by the 2080s, up to 330 million people would be at risk from malaria`. This is a substantial increase compared to the 1920s, where the number at risk was between 50 to 100 million.\nThis is correctly referred from this table in the PDF!\n![](../images/body-9afc8c92de35e1eba073.png)\nSource: Page 3 of [The Impacts and Costs of Climate Change](https://climate.ec.europa.eu/system/files/2016-11/final_report2_en.pdf)\n**#2. Disassociation between texts and images**\n**Question** : What’s the modelled cost of climate change, according to the Greenbook declining model?\n**Answer** : According to the Greenbook declining model, the modelled cost of climate change is 7.2 Euro/tCO2\nThe correct answer should be 7.4 Euro/tCO2, but we can see it is close to correct!\n![](../images/body-1febdf415bb1a3e01f71.png)\nSource: Page 46 of [The Impacts and Costs of Climate Change](https://climate.ec.europa.eu/system/files/2016-11/final_report2_en.pdf)\n## Conclusion\nTraditional RAG pipelines struggle with non-textual content. ColPali treats each PDF page as an image, allowing it to process visual layouts, tables, charts, and embedded graphics — formats that standard text parsers often distort or ignore.\nColPali brings vision-language intelligence to RAG, making it far more capable of handling the messy, multimodal reality of enterprise documents.\n_*Unless otherwise noted, all images are by the author_\n## Register Now to Continue Reading\n### Unlock Towards Data Science with a Free Account\nAll articles on Towards Data Science are available to registered users. Create a free account today to access this article and thousands more.\nRegister for freeSign in to your account\n* * *\n![](../images/body-d6f10b7f094af8ff005c.jpeg)\nWritten By\n## [Julian Yip](https://towardsdatascience.com/author/jyipkl/)\n[See all from Julian Yip](https://towardsdatascience.com/author/jyipkl/)\n[Editor's Picks](https://towardsdatascience.com/tag/editors-pick/)[Artificial Intelligence](https://towardsdatascience.com/tag/artificial-intelligence/)[Generative Ai](https://towardsdatascience.com/tag/generative-ai/)[Multimodality](https://towardsdatascience.com/tag/multimodality/)[Rag](https://towardsdatascience.com/tag/rag/)\nShare This Article\n * [](https://www.facebook.com/sharer/sharer.php?u=https%3A%2F%2Ftowardsdatascience.com%2Fbringing-vision-language-intelligence-to-rag-with-colpali%2F&title=Bringing%20Vision-Language%20Intelligence%20to%20RAG%20with%20ColPali)\n * [](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Ftowardsdatascience.com%2Fbringing-vision-language-intelligence-to-rag-with-colpali%2F&title=Bringing%20Vision-Language%20Intelligence%20to%20RAG%20with%20ColPali)\n * [](https://x.com/share?url=https%3A%2F%2Ftowardsdatascience.com%2Fbringing-vision-language-intelligence-to-rag-with-colpali%2F&text=Bringing%20Vision-Language%20Intelligence%20to%20RAG%20with%20ColPali)\n\n\nTowards Data Science is a community publication. Submit your insights to reach our global audience and earn through the TDS Author Payment Program.\n[Write for TDS](https://towardsdatascience.com/submissions/)\n## Related Articles\n * [![](../images/003.jpg)](https://towardsdatascience.com/detecting-insecure-code-with-llms-8b8ad923dd98/)\n## [Detecting Insecure Code with LLMs](https://towardsdatascience.com/detecting-insecure-code-with-llms-8b8ad923dd98/)\n[Large Language Models](https://towardsdatascience.com/category/artificial-intelligence/large-language-models/)\nPrompt Experiments for Python Vulnerability Detection\n[Melanie Hart Buehler](https://towardsdatascience.com/author/melanie.h.buehler/)\nMarch 21, 202411 min read\n * [![](../images/004.jpeg)](https://towardsdatascience.com/running-llama-2-on-cpu-inference-for-document-q-a-3d636037a3d8/)\n## [Running Llama 2 on CPU Inference Locally for Document Q&A](https://towardsdatascience.com/running-llama-2-on-cpu-inference-for-document-q-a-3d636037a3d8/)\n[Large Language Models](https://towardsdatascience.com/category/artificial-intelligence/large-language-models/)\nClearly explained guide for running quantized open-source LLM applications on CPUs using LLama 2, C Transformers, GGML, and LangChain\n[Kenneth Leung](https://towardsdatascience.com/author/kennethleungty/)\nJuly 18, 20234 min read\n * [![](../images/005.jpeg)](https://towardsdatascience.com/benchmarking-time-series-datasets-with-style-2a4e8789a965/)\n## [Benchmarking time series datasets with style](https://towardsdatascience.com/benchmarking-time-series-datasets-with-style-2a4e8789a965/)\n[](https://towardsdatascience.com/bringing-vision-language-intelligence-to-rag-with-colpali/)\nHow to benchmark/spot differences between time series datasets?\n[M Adel](https://towardsdatascience.com/author/muadelm/)\nOctober 20, 20213 min read\n * [![](../images/006.png)](https://towardsdatascience.com/topic-model-labelling-with-llms/)\n## [Topic Model Labelling with LLMs](https://towardsdatascience.com/topic-model-labelling-with-llms/)\n[Large Language Models](https://towardsdatascience.com/category/artificial-intelligence/large-language-models/)\nPython tutorial for reproducible labeling of cutting-edge topic models with GPT4-o-mini.\n[Petr Koráb](https://towardsdatascience.com/author/petrkorab/)\nJuly 14, 20256 min read\n * [![](../images/007.png)](https://towardsdatascience.com/highlights-on-large-language-models-at-kdd-2023-fc53440563c3/)\n## [Highlights on Large Language Models at KDD 2023](https://towardsdatascience.com/highlights-on-large-language-models-at-kdd-2023-fc53440563c3/)\n[Deep Learning](https://towardsdatascience.com/category/artificial-intelligence/deep-learning/)\nWere you not able to attend KDD? Learn from my summary about the hottest topic at the conference: LLMs\n[Gabriel Moreira](https://towardsdatascience.com/author/gspmoreira/)\nSeptember 11, 20238 min read\n * [![](../images/008.jpeg)](https://towardsdatascience.com/heart-disease-prediction-73468d630cfc/)\n## [Heart Disease Prediction](https://towardsdatascience.com/heart-disease-prediction-73468d630cfc/)\n[Artificial Intelligence](https://towardsdatascience.com/category/artificial-intelligence/)\nCleveland Heart Disease(UCI Repository) dataset - classification with various models.\n[Shubhankar Rawat](https://towardsdatascience.com/author/shubhankarrawatsam.1999/)\nAugust 10, 20198 min read\n * [![](../images/009.jpg)](https://towardsdatascience.com/i-invented-a-way-to-speak-to-an-ai-keeping-your-privacy-ddbca5f24e4a/)\n## [I Invented a Way to Speak to an AI, Keeping Your Privacy](https://towardsdatascience.com/i-invented-a-way-to-speak-to-an-ai-keeping-your-privacy-ddbca5f24e4a/)\n[Artificial Intelligence](https://towardsdatascience.com/category/artificial-intelligence/)\nThe tech is called \"Silent Voice.\"\n[Rafe Brena, Ph.D.](https://towardsdatascience.com/author/rafebrena/)\nJune 27, 20248 min read\n * [![](../images/010.jpg)](https://towardsdatascience.com/how-to-enrich-llm-context-to-significantly-enhance-capabilities/)\n## [How to Enrich LLM Context to Significantly Enhance Capabilities](https://towardsdatascience.com/how-to-enrich-llm-context-to-significantly-enhance-capabilities/)\n[Large Language Models](https://towardsdatascience.com/category/artificial-intelligence/large-language-models/)\nLearn how to empower your LLMs by leveraging additional metadata\n[Eivind Kjosbakken](https://towardsdatascience.com/author/oieivind/)\nSeptember 16, 20257 min read\n * [![](../images/011.jpg)](https://towardsdatascience.com/text-watermarking-in-python-catch-whoever-copies-your-writing/)\n## [Text Watermarking in Python: Catch Whoever Copies Your Writing](https://towardsdatascience.com/text-watermarking-in-python-catch-whoever-copies-your-writing/)\n[Large Language Models](https://towardsdatascience.com/category/artificial-intelligence/large-language-models/)\nAI companies quietly watermark billions of words a day. Here’s how to apply the same three families of techniques to your own writing—and what real experiments reveal about which watermarks survive copy-paste, editing, and paraphrasing.\n[Chien Vu Minh](https://towardsdatascience.com/author/chienvu/)\nSeptember 6, 202626 min read\n\n\n * [YouTube](https://www.youtube.com/c/TowardsDataScience)\n * [X](https://x.com/TDataScience)\n * [LinkedIn](https://www.linkedin.com/company/towards-data-science/)\n * [Threads](https://www.threads.net/@towardsdatascience)\n * [Bluesky](https://bsky.app/profile/towardsdatascience.com)\n\n\n[](https://towardsdatascience.com/)\nYour home for data science and AI. The world's leading publication for data science, data analytics, data engineering, machine learning, and artificial intelligence professionals.\nSubscribe to Our Newsletter[Write For TDS](https://towardsdatascience.com/submissions/)•[About](https://towardsdatascience.com/about-towards-data-science-d691af11cc2f/)•[Advertise](https://contact.towardsdatascience.com/advertise-with-towards-data-science)•[Privacy Policy](https://towardsdatascience.com/privacy-policy/)•[Terms of Use](https://towardsdatascience.com/website-terms-of-use/)\n© Insight Media Group, LLC 2026•[The New Stack](https://thenewstack.io)•[roadmap.sh](https://roadmap.sh)•[Creator Network](https://insightmediagroup.io/creator-network)\n#### Recommended Articles \nClose\n * [ ![](https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/607686/image?etag=1782161464448&width=500&height=500) How Relevance Models Foreshadowed Transformers for NLP ](https://towardsdatascience.com/how-relevance-models-foreshadowed-transformers-for-nlp/ \"How Relevance Models Foreshadowed Transformers for NLP\")\n * [ ![](https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/607639/image?etag=1781250125848&width=500&height=500) Javascript Fatigue: HTMX Is All You Need to Build ChatGPT — Part 1 ](https://towardsdatascience.com/javascript-fatigue-you-dont-need-js-to-build-chatgpt/ \"Javascript Fatigue: HTMX Is All You Need to Build ChatGPT — Part 1\")\n * [ ![](https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/607930/image?etag=1782124566054&width=500&height=500) The Subset Sum Problem Solved in Linear Time for Dense Enough Inputs ](https://towardsdatascience.com/subset-sum-problem-solved-in-linear-time-for-dense-enough-inputs/ \"The Subset Sum Problem Solved in Linear Time for Dense Enough Inputs\")\n * [ ![](https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/3660/image?etag=1780176246070&width=500&height=500) Does Your Company Have a Data Strategy? ](https://towardsdatascience.com/does-your-company-have-a-data-strategy-e8944dbec4ae/ \"Does Your Company Have a Data Strategy?\")\n\n\n", "metadata": { "title": "Bringing Vision-Language Intelligence to RAG with ColPali | Towards Data Science", "description": "Unlocking the value of non-textual contents in your knowledge base", "keywords": null, "author": "Julian Yip", "og:locale": "en_US", "og:type": "article", "og:title": "Bringing Vision-Language Intelligence to RAG with ColPali", "og:description": "Unlocking the value of non-textual contents in your knowledge base", "og:url": "https://towardsdatascience.com/bringing-vision-language-intelligence-to-rag-with-colpali/", "og:site_name": "Towards Data Science", "og:image": "https://assets.insightmediagroup.io/media/wp-content/uploads/2025/10/image-407.png", "og:image:width": "576", "og:image:height": "384", "og:image:type": "image/png", "twitter:card": "summary_large_image", "twitter:site": "@TDataScience", "twitter:creator": "@TDataScience", "twitter:label1": "Written by", "twitter:data1": "Julian Yip", "twitter:label2": "Est. reading time", "twitter:data2": "7 min read", "article:published_time": "2025-10-29T12:53:40.000Z", "article:modified_time": "2025-10-29T12:53:40.000Z", "article:author": "https://towardsdatascience.com/author/jyipkl/" }, "paywall": { "listed_paid": false, "detected": false, "marker": "", "public_only": false, "policy": "付费/会员内容只保留页面公开可见部分，不绕过付费墙" }, "crawl": { "success": true, "status_code": 202, "error": "" }, "code_blocks": [], "images": [ { "status": "http_404", "url": "https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/607686/image?etag=1782161464448&width=500&height=500", "src": "https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/607686/image?etag=1782161464448&width=500&height=500", "alt": "", "downloaded": false, "discovery": "saved_markdown", "retried_at": "2026-10-20T17:38:36Z" }, { "status": "http_404", "url": "https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/607639/image?etag=1781250125848&width=500&height=500", "src": "https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/607639/image?etag=1781250125848&width=500&height=500", "alt": "", "downloaded": false, "discovery": "saved_markdown", "retried_at": "2026-10-20T17:38:36Z" }, { "status": "http_404", "url": "https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/607930/image?etag=1782124566054&width=500&height=500", "src": "https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/607930/image?etag=1782124566054&width=500&height=500", "alt": "", "downloaded": false, "discovery": "saved_markdown", "retried_at": "2026-10-20T17:38:36Z" }, { "status": "http_404", "url": "https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/3660/image?etag=1780176246070&width=500&height=500", "src": "https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/3660/image?etag=1780176246070&width=500&height=500", "alt": "", "downloaded": false, "discovery": "saved_markdown", "retried_at": "2026-10-20T17:38:36Z" } ], "audio": [], "videos": [ { "url": "https://www.youtube.com/c/TowardsDataScience", "kind": "channel_reference", "status": "reference_only" } ], "links": [ { "type": "internal", "href": "https://towardsdatascience.com/" }, { "type": "internal", "href": "https://towardsdatascience.com/latest/" }, { "type": "internal", "href": "https://towardsdatascience.com/tag/editors-pick/" }, { "type": "internal", "href": "https://towardsdatascience.com/tag/the-variable/" }, { "type": "internal", "href": "https://towardsdatascience.com/shipai/" }, { "type": "internal", "href": "https://towardsdatascience.com/submissions/" }, { "type": "internal", "href": "https://towardsdatascience.com/category/artificial-intelligence/large-language-models/" }, { "type": "internal", "href": "https://towardsdatascience.com/author/jyipkl/" }, { "type": "internal", "href": "https://towardsdatascience.com/tag/artificial-intelligence/" }, { "type": "internal", "href": "https://towardsdatascience.com/tag/generative-ai/" }, { "type": "internal", "href": "https://towardsdatascience.com/tag/multimodality/" }, { "type": "internal", "href": "https://towardsdatascience.com/tag/rag/" }, { "type": "internal", "href": "https://towardsdatascience.com/detecting-insecure-code-with-llms-8b8ad923dd98/" }, { "type": "internal", "href": "https://towardsdatascience.com/author/melanie.h.buehler/" }, { "type": "internal", "href": "https://towardsdatascience.com/running-llama-2-on-cpu-inference-for-document-q-a-3d636037a3d8/" }, { "type": "internal", "href": "https://towardsdatascience.com/author/kennethleungty/" }, { "type": "internal", "href": "https://towardsdatascience.com/benchmarking-time-series-datasets-with-style-2a4e8789a965/" }, { "type": "internal", "href": "https://towardsdatascience.com/bringing-vision-language-intelligence-to-rag-with-colpali/" }, { "type": "internal", "href": "https://towardsdatascience.com/author/muadelm/" }, { "type": "internal", "href": "https://towardsdatascience.com/topic-model-labelling-with-llms/" }, { "type": "external", "href": "https://write.insightmediagroup.io" }, { "type": "external", "href": "https://www.linkedin.com/company/towards-data-science/" }, { "type": "external", "href": "https://x.com/TDataScience" }, { "type": "external", "href": "https://roadmap.sh/ai-engineer" }, { "type": "external", "href": "https://climate.ec.europa.eu/system/files/2016-11/final_report2_en.pdf" }, { "type": "external", "href": "https://github.com/yip-kl?tab=repositories" }, { "type": "external", "href": "https://huggingface.co/spaces/vidore/vidore-leaderboard" }, { "type": "external", "href": "https://www.facebook.com/sharer/sharer.php?title=Bringing+Vision-Language+Intelligence+to+RAG+with+ColPali&u=https%3A%2F%2Ftowardsdatascience.com%2Fbringing-vision-language-intelligence-to-rag-with-colpali%2F" }, { "type": "external", "href": "https://www.linkedin.com/shareArticle?mini=true&title=Bringing+Vision-Language+Intelligence+to+RAG+with+ColPali&url=https%3A%2F%2Ftowardsdatascience.com%2Fbringing-vision-language-intelligence-to-rag-with-colpali%2F" }, { "type": "external", "href": "https://x.com/share?text=Bringing+Vision-Language+Intelligence+to+RAG+with+ColPali&url=https%3A%2F%2Ftowardsdatascience.com%2Fbringing-vision-language-intelligence-to-rag-with-colpali%2F" }, { "type": "external", "href": "https://www.youtube.com/c/TowardsDataScience" }, { "type": "external", "href": "https://www.threads.net/@towardsdatascience" }, { "type": "external", "href": "https://bsky.app/profile/towardsdatascience.com" }, { "type": "external", "href": "https://thenewstack.io" }, { "type": "external", "href": "https://roadmap.sh" }, { "type": "external", "href": "https://insightmediagroup.io/creator-network" } ], "repair_version": 2, "repaired_at": "2026-10-20T17:24:02Z", "source_url": "https://towardsdatascience.com/bringing-vision-language-intelligence-to-rag-with-colpali/", "delivery_status": "unverified", "discovery_status": "http_403", "missing_reason": "源页面未重新核验:http_403；已按保留正文补采图片", "repair_error": "", "actual_video_files": [], "expected_saved_body_images": 4, "image_discovery": "saved_markdown", "source_body_markdown": "![Revisit consent button](https://cdn-cookieyes.com/assets/images/revisit.svg)\nWe value your privacy\nWe use cookies to enhance your browsing experience, serve personalised ads or content, and analyse our traffic. By clicking \"Accept All\", you consent to our use of cookies.\nCustomise Reject All Accept All\nCustomise Consent Preferences ![](https://cdn-cookieyes.com/assets/images/close.svg)\nWe use cookies to help you navigate efficiently and perform certain functions. You will find detailed information about all cookies under each consent category below.\nThe cookies that are categorised as \"Necessary\" are stored on your browser as they are essential for enabling the basic functionalities of the site. ... Show more\nNecessaryAlways Active\nNecessary cookies are required to enable the basic features of this site, such as providing secure log-in or adjusting your consent preferences. These cookies do not store any personally identifiable data.\n * Cookie\nBCTempID\n * Duration\n10 minutes\n * Description\nNo description available.\n\n\n * Cookie\n__cf_bm\n * Duration\n1 hour\n * Description\nThis cookie, set by Cloudflare, is used to support Cloudflare Bot Management. \n\n\n * Cookie\nAWSALBCORS\n * Duration\n7 days\n * Description\nAmazon Web Services set this cookie for load balancing.\n\n\n * Cookie\n_cfuvid\n * Duration\nsession\n * Description\nCloudflare sets this cookie to track users across sessions to optimize user experience by maintaining session consistency and providing personalized services\n\n\n * Cookie\nli_gc\n * Duration\n6 months\n * Description\nLinkedin set this cookie for storing visitor's consent regarding using cookies for non-essential purposes.\n\n\n * Cookie\n__hssrc\n * Duration\nsession\n * Description\nThis cookie is set by Hubspot whenever it changes the session cookie. The __hssrc cookie set to 1 indicates that the user has restarted the browser, and if the cookie does not exist, it is assumed to be a new session.\n\n\n * Cookie\n__hssc\n * Duration\n1 hour\n * Description\nHubSpot sets this cookie to keep track of sessions and to determine if HubSpot should increment the session number and timestamps in the __hstc cookie.\n\n\n * Cookie\nwpEmojiSettingsSupports\n * Duration\nsession\n * Description\nWordPress sets this cookie when a user interacts with emojis on a WordPress site. It helps determine if the user's browser can display emojis properly.\n\n\n * Cookie\nBCSessionID\n * Duration\n1 year 1 month 4 days\n * Description\nBlueconic sets this cookie as a unique identifier for the BlueConic profile.\n\n\n * Cookie\n_octo\n * Duration\n1 year\n * Description\nNo description available.\n\n\n * Cookie\nlogged_in\n * Duration\n1 year\n * Description\nNo description available.\n\n\n * Cookie\n__Secure-YEC\n * Duration\npast\n * Description\nYouTube sets this cookie to stores the user's video player preferences using embedded YouTube video\n\n\n * Cookie\n__eoi\n * Duration\n6 months\n * Description\nDescription is currently not available.\n\n\n * Cookie\nAWSALBTGCORS\n * Duration\n7 days\n * Description\nNo description available.\n\n\n * Cookie\nlogin-status-p\n * Duration\npast\n * Description\nDescription is currently not available.\n\n\n * Cookie\nAWSALBTG\n * Duration\n7 days\n * Description\nNo description available.\n\n\n * Cookie\ncsrf_token\n * Duration\nsession\n * Description\nNo description available.\n\n\n * Cookie\ntoken_v2\n * Duration\n1 day\n * Description\nDescription is currently not available.\n\n\n * Cookie\nD\n * Duration\n1 year\n * Description\nDescription is currently not available.\n\n\n * Cookie\nPHPSESSID\n * Duration\nsession\n * Description\nThis cookie is native to PHP applications. The cookie stores and identifies a user's unique session ID to manage user sessions on the website. The cookie is a session cookie and will be deleted when all the browser windows are closed.\n\n\n * Cookie\nVISITOR_PRIVACY_METADATA\n * Duration\n6 months\n * Description\nYouTube sets this cookie to store the user's cookie consent state for the current domain. \n\n\n * Cookie\ncookietest\n * Duration\nsession\n * Description\nThe cookietest cookie is typically used to determine whether the user's browser accepts cookies, essential for website functionality and user experience.\n\n\n * Cookie\n__Host-airtable-session\n * Duration\n1 year\n * Description\nThis cookie is used to enable us to integrate the services of Airtable.\n\n\n * Cookie\n__Host-airtable-session.sig\n * Duration\n1 year\n * Description\nThis cookie is used to enable us to integrate the services of Airtable.\n\n\n * Cookie\nm\n * Duration\n1 year 1 month 4 days\n * Description\nStripe sets this cookie for fraud prevention purposes. It identifies the device used to access the website, allowing the website to be formatted accordingly.\n\n\n * Cookie\nBIGipServer*\n * Duration\nsession\n * Description\nMarketo sets this cookie to collect information about the user's online activity and build a profile about their interests to provide advertisements relevant to the user. \n\n\n * Cookie\n__cfruid\n * Duration\nsession\n * Description\nCloudflare sets this cookie to identify trusted web traffic.\n\n\n * Cookie\n_GRECAPTCHA\n * Duration\n6 months\n * Description\nGoogle Recaptcha service sets this cookie to identify bots to protect the website against malicious spam attacks.\n\n\n * Cookie\n__Secure-YNID\n * Duration\n6 months\n * Description\nGoogle cookie used to protect user security and prevent fraud, especially during the login process.\n\n\n * Cookie\ncookieyes-consent\n * Duration\n1 year\n * Description\nCookieYes sets this cookie to remember users' consent preferences so that their preferences are respected on subsequent visits to this site. It does not collect or store any personal information about the site visitors.\n\n\nFunctional\nFunctional cookies help perform certain functionalities like sharing the content of the website on social media platforms, collecting feedback, and other third-party features.\n * Cookie\nlidc\n * Duration\n1 day\n * Description\nLinkedIn sets the lidc cookie to facilitate data center selection.\n\n\n * Cookie\nbrw\n * Duration\n1 year\n * Description\nNo description available.\n\n\n * Cookie\nbrwConsent\n * Duration\n5 minutes\n * Description\nDescription is currently not available.\n\n\n * Cookie\nWMF-Uniq\n * Duration\n1 year\n * Description\nDescription is currently not available.\n\n\n * Cookie\nloom_anon_comment\n * Duration\n1 year\n * Description\nNo description available.\n\n\n * Cookie\nloom_referral_video\n * Duration\nsession\n * Description\nDescription is currently not available.\n\n\n * Cookie\nVISITOR_INFO1_LIVE\n * Duration\n6 months\n * Description\nA cookie set by YouTube to measure bandwidth that determines whether the user gets the new or old player interface.\n\n\n * Cookie\nyt-remote-connected-devices\n * Duration\nNever Expires\n * Description\nYouTube sets this cookie to store the user's video preferences using embedded YouTube videos.\n\n\n * Cookie\nytidb::LAST_RESULT_ENTRY_KEY\n * Duration\nNever Expires\n * Description\nThe cookie ytidb::LAST_RESULT_ENTRY_KEY is used by YouTube to store the last search result entry that was clicked by the user. This information is used to improve the user experience by providing more relevant search results in the future.\n\n\n * Cookie\nyt-remote-device-id\n * Duration\nNever Expires\n * Description\nYouTube sets this cookie to store the user's video preferences using embedded YouTube videos.\n\n\n * Cookie\nyt-remote-session-name\n * Duration\nsession\n * Description\nThe yt-remote-session-name cookie is used by YouTube to store the user's video player preferences using embedded YouTube video.\n\n\n * Cookie\nyt-remote-fast-check-period\n * Duration\nsession\n * Description\nThe yt-remote-fast-check-period cookie is used by YouTube to store the user's video player preferences for embedded YouTube videos.\n\n\n * Cookie\nyt-remote-session-app\n * Duration\nsession\n * Description\nThe yt-remote-session-app cookie is used by YouTube to store user preferences and information about the interface of the embedded YouTube video player.\n\n\n * Cookie\nyt-remote-cast-available\n * Duration\nsession\n * Description\nThe yt-remote-cast-available cookie is used to store the user's preferences regarding whether casting is available on their YouTube video player.\n\n\n * Cookie\nyt-remote-cast-installed\n * Duration\nsession\n * Description\nThe yt-remote-cast-installed cookie is used to store the user's video player preferences using embedded YouTube video.\n\n\n * Cookie\ncp_session\n * Duration\n3 months\n * Description\nCodepen sets this cookie for Help systems found in the website.\n\n\n * Cookie\nloid\n * Duration\n1 year 1 month 4 days\n * Description\nThis cookie is set by the Reddit. The cookie enables the sharing of content from the website onto the social media platform.\n\n\nAnalytics\nAnalytical cookies are used to understand how visitors interact with the website. These cookies help provide information on metrics such as the number of visitors, bounce rate, traffic source, etc.\n * Cookie\n__hstc\n * Duration\n6 months\n * Description\nHubspot set this main cookie for tracking visitors. It contains the domain, initial timestamp (first visit), last timestamp (last visit), current timestamp (this visit), and session number (increments for each subsequent session).\n\n\n * Cookie\nhubspotutk\n * Duration\n6 months\n * Description\nHubSpot sets this cookie to keep track of the visitors to the website. This cookie is passed to HubSpot on form submission and used when deduplicating contacts.\n\n\n * Cookie\n_ga\n * Duration\n1 year 1 month 4 days\n * Description\nGoogle Analytics sets this cookie to calculate visitor, session and campaign data and track site usage for the site's analytics report. The cookie stores information anonymously and assigns a randomly generated number to recognise unique visitors.\n\n\n * Cookie\n_ga_*\n * Duration\n1 year 1 month 4 days\n * Description\nGoogle Analytics sets this cookie to store and count page views.\n\n\n * Cookie\n__Host-psifi.analyticsTrace\n * Duration\n6 hours\n * Description\nDescription is currently not available.\n\n\n * Cookie\n__Host-psifi.analyticsTraceV2\n * Duration\n6 hours\n * Description\nDescription is currently not available.\n\n\n * Cookie\n_gh_sess\n * Duration\nsession\n * Description\nGitHub sets this cookie for temporary application and framework state between pages like what step the user is on in a multiple step form.\n\n\n * Cookie\nYSC\n * Duration\nsession\n * Description\nYSC cookie is set by Youtube and is used to track the views of embedded videos on Youtube pages.\n\n\n * Cookie\najs_anonymous_id\n * Duration\n1 year\n * Description\nThis cookie is set by Segment to count the number of people who visit a certain site by tracking if they have visited before.\n\n\n * Cookie\nvuid\n * Duration\n1 year 1 month 4 days\n * Description\nVimeo installs this cookie to collect tracking information by setting a unique ID to embed videos on the website. \n\n\nPerformance\nPerformance cookies are used to understand and analyse the key performance indexes of the website which helps in delivering a better user experience for the visitors.\n * Cookie\nAWSALB\n * Duration\n7 days\n * Description\nAWSALB is an application load balancer cookie set by Amazon Web Services to map the session to the target.\n\n\n * Cookie\nacq\n * Duration\npast\n * Description\nDescription is currently not available.\n\n\n * Cookie\nacq.sig\n * Duration\npast\n * Description\nDescription is currently not available.\n\n\n * Cookie\nptc\n * Duration\n2 years\n * Description\nNo description available.\n\n\nAdvertisement\nAdvertisement cookies are used to provide visitors with customised advertisements based on the pages you visited previously and to analyse the effectiveness of the ad campaigns.\n * Cookie\nmuc_ads\n * Duration\n1 year 1 month 4 days\n * Description\nTwitter sets this cookie to collect user behaviour and interaction data to optimize the website.\n\n\n * Cookie\nguest_id_marketing\n * Duration\n1 year 1 month 4 days\n * Description\nTwitter sets this cookie to identify and track the website visitor.\n\n\n * Cookie\nguest_id_ads\n * Duration\n1 year 1 month 4 days\n * Description\nTwitter sets this cookie to identify and track the website visitor.\n\n\n * Cookie\npersonalization_id\n * Duration\n1 year 1 month 4 days\n * Description\nTwitter sets this cookie to integrate and share features for social media and also store information about how the user uses the website, for tracking and targeting.\n\n\n * Cookie\nguest_id\n * Duration\n1 year 1 month 4 days\n * Description\nTwitter sets this cookie to identify and track the website visitor. It registers if a user is signed in to the Twitter platform and collects information about ad preferences.\n\n\n * Cookie\nbcookie\n * Duration\n1 year\n * Description\nLinkedIn sets this cookie from LinkedIn share buttons and ad tags to recognize browser IDs.\n\n\n * Cookie\n__Secure-ROLLOUT_TOKEN\n * Duration\n6 months\n * Description\nYouTube sets this cookie to manage feature rollout and experimentation. It helps Google control which new features or interface changes are shown to users as part of testing and staged rollouts, ensuring consistent experience for a given user during an experiment.\n\n\n * Cookie\nyt.innertube::nextId\n * Duration\nNever Expires\n * Description\nYouTube sets this cookie to register a unique ID to store data on what videos from YouTube the user has seen.\n\n\n * Cookie\nyt.innertube::requests\n * Duration\nNever Expires\n * Description\nYouTube sets this cookie to register a unique ID to store data on what videos from YouTube the user has seen.\n\n\n * Cookie\nsession_tracker\n * Duration\nsession\n * Description\nThis cookie is set by the Reddit. This cookie is used to identify trusted web traffic. It also helps in adverstising on the website.\n\n\n * Cookie\nedgebucket\n * Duration\nsession\n * Description\nReddit sets this cookie to save the information about a log-on Reddit user, for the purpose of advertisement recommendations and updating the content.\n\n\n * Cookie\ndid\n * Duration\n1 year\n * Description\nArbor sets this cookie to show targeted ads to site visitors.This cookie expires after 2 months or 1 year.\n\n\nUncategorised\nOther uncategorised cookies are those that are being analysed and have not been classified into a category as yet.\nNo cookies to display.\nReject All Save My Preferences Accept All\nToggle Search[](https://towardsdatascience.com/)Toggle Mobile Navigation\n[](https://towardsdatascience.com/)\nPublish AI, ML & data-science insights to a global community of data professionals.\nSign in[Submit an Article](https://write.insightmediagroup.io)\n[Latest](https://towardsdatascience.com/latest/)[Editor's Picks](https://towardsdatascience.com/tag/editors-pick/)[Newsletter](https://towardsdatascience.com/tag/the-variable/)[ShipAI](https://towardsdatascience.com/shipai/)\n* * *\n[Write For TDS](https://towardsdatascience.com/submissions/)\n[LinkedIn](https://www.linkedin.com/company/towards-data-science/)[X](https://x.com/TDataScience)Toggle Search\n[Large Language Models](https://towardsdatascience.com/category/artificial-intelligence/large-language-models/)\n# Bringing Vision-Language Intelligence to RAG with ColPali\nUnlocking the value of non-textual contents in your knowledge base\n[Julian Yip](https://towardsdatascience.com/author/jyipkl/)\nOctober 29, 20257 min read\n![](https://assets.insightmediagroup.io/media/wp-content/uploads/2025/10/image-407.png)\nSource: Image by the author.\nIf you’ve ever tried building a RAG (Retrieval-Augmented Generation) application, you’re likely familiar with the challenges posed by tables and images. This article explores how to tackle these formats using Vision Language Models, specifically with the ColPali model.\nBut first, what exactly is RAG — and why do tables and images make it so difficult?\n## RAG and parsing\nImagine you’re faced with a question like:\n> What’s the our company’s policy for handling refund?\nA foundational LLM (Large Language Model) probably won’t be able to answer this, as such information is company-specific and typically not included in the model’s training data.\nLearn this step by step with the interactive [AI Engineer](https://roadmap.sh/ai-engineer) roadmap.\nThat’s why a common approach is to connect the LLM to a knowledge base — such as a SharePoint folder containing various internal documents. This allows the model to retrieve and incorporate relevant context, enabling it to answer questions that require specialized knowledge. This technique is known as Retrieval-Augmented Generation (RAG), and it often involves working with documents like PDFs.\nHowever, extracting the right information from a large and diverse knowledge base requires extensive document preprocessing. Common steps include:\n 1. **Parsing** : Parsing documents into texts and images, often assisted with Optical Character Recognition (OCR) tools like Tesseract. Tables are most often converted into texts\n 2. **Structure Preservation** : Maintain the structure of the document, including headings, paragraphs, by converting the extracted text into a format that retains context, such as Markdown\n 3. **Chunking** : Splitting or merging text passages, so that the contexts can be fed into the context window without causing the passages come across as disjointed\n 4. **Enriching** : Provide extra metadata e.g. extract keyword or provide summary to the chunks to ease discovery. Optionally, to also caption images with descriptive texts via multimodal LLM to make images searchable\n 5. **Embedding** : Embed the texts (and potentially the images too with multimodal embedding), and store them into a vector DB\n\n\nAs you can imagine, the process is highly complicated, involves a lot of experimentation, and is very brittle. Worse yet, even if we tried to do it as best as we could, this parsing might not actually work after all.\n## Why parsing often falls short\nTables and image often exist in PDFs. The below image shows how they are typically parsed for LLM’s consumption:\n![](https://assets.insightmediagroup.io/media/wp-content/uploads/2025/10/1cl5ttrnhf4XACGc_nM9Vxw.png)\nSource: Image by the author.\n * **Texts** are chunked\n * **Tables** are turned into texts, whatever contained within are copied without preserving table boundaries\n * **Images** are fed into multimodal LLM for text summary generation, or alternatively, the original image is fed into multimodal embedding model without needing to generate a text summary\n\n\nHowever, there are two inherent issues with such traditional approach.\n**#1. Complex tables cannot be simply be interpreted as texts** \nTaking this table as an example, we as human would interpret that a temperature change of **> 2˚C to 2.5˚C’**s implication on **Health** is **A rise of 2.3˚C by 2080 puts up to 270 million at risk from malaria**\n![](https://assets.insightmediagroup.io/media/wp-content/uploads/2025/10/1rpLlAcPctSz7x6UU_XFRyg.png)\nSource: [The Impacts and Costs of Climate Change](https://climate.ec.europa.eu/system/files/2016-11/final_report2_en.pdf)\nHowever, if we turn this table into a text, it would look like this: `Temperature change Within EC target <(2˚C) >2˚C to 2.5˚C >3C Health Globally it is estimated that A rise of 2.3oC by 2080 puts A rise of 3.3oC by 2080 an average temperature rise up to 270 million at risk from would put up to 330...`\nThe result is a jumbled block of text with no discernible meaning. Even for a human reader, it’s impossible to extract any meaningful insight from it. When this kind of text is fed into a Large Language Model (LLM), it also fails to produce an accurate interpretation.\n**#2. Disassociation between texts and images** \nThe description of the image is often included in texts and they are inseparable from one another. Taking the below as an example, we know the chart represents the “Modelled Costs of Climate Change with Different Pure Rate of Time Preference and declining discount rate schemes (no equity weighting)”\n![](https://assets.insightmediagroup.io/media/wp-content/uploads/2025/10/1XTGA9EL704RE05tIZlZ42w.png)\nSource: [The Impacts and Costs of Climate Change](https://climate.ec.europa.eu/system/files/2016-11/final_report2_en.pdf)\nHowever, as this is parsed, the image description (parsed text) will be disassociated with the image (parsed chart). So we can expect, during RAG, the image would not be retrieved as input when we raise a question like “what’s the cost of climate change?”\n![](https://assets.insightmediagroup.io/media/wp-content/uploads/2025/10/1_bdHYj4bp9vGah4elDoXLQ.png)\nSo, even if we attempt to engineer solutions that preserve as much information as possible during parsing, they often fall short when faced with real-world scenarios.\nGiven how critical parsing is in RAG applications, does this mean RAG agents are destined to fail when working with complex documents? Absolutely not. With ColPali, we have a more refined and effective approach to handling them.\n## What’s ColPali?\nThe core premise of ColPali is simple: Human read PDF as pages, not “chunks”, so it makes sense to treat PDF as such: Instead of going through the messy process of parsing, we just turn the PDF pages into images, and use that as context for LLM to provide an answer.\n![](https://assets.insightmediagroup.io/media/wp-content/uploads/2025/10/1Cot_klZZgwluKfZamNBRag.png)\nNow, the idea of embedding images using multimodal models isn’t new — it’s a common technique. So what makes ColPali stand out? The key lies in its inspiration from ColBERT, a model that embeds inputs into multi-vectors, enabling more precise and efficient search.\nBefore diving into ColPali’s capabilities, let me briefly digress to explain what ColBERT is all about.\n#### ColBERT: Granular, context-aware embedding for texts\nColBERT is a text embedding and reranking technique that leverage on multi-vectors to enhance search accuracy for texts.\nLet’s consider this case: we have this question: `is Paul vegan?`, we need to identify which text chuck contains the relevant information.\n![](https://assets.insightmediagroup.io/media/wp-content/uploads/2025/10/1LWeoTQNYShSUQXtV6Jzf4w.png)\n_Highlighted in yellow are texts which contain information about Paul_\nIdeally, we should identify `Text Chunk A` as the most relevant one. But if we use a single-vector embedding model (text-ada-002), it will return `Text Chunk B` instead.\nThe reason lies in how single-vector bi-encoders — like text-ada-002 — operate. They attempt to compress an entire sentence into a single vector, without encoding individual words in a context-aware manner. In contrast, ColBERT **embeds each word with contextual awareness** , resulting in a richer, **multi-vector** representation that captures more nuanced information.\n![](https://assets.insightmediagroup.io/media/wp-content/uploads/2025/10/15YoRcP42Tgu-QMvOvTgCcA.png)\nNumbers in the vectors are illustrative and do not represents the actual values\n#### ColPali: ColBERT’s brother for handling document-like images\nColPali follows a similar philosophy but applies it to document-like images. Just as ColBERT breaks down text and embeds each word individually, ColPali divides an image into patches and generates embeddings for each patch. This approach preserves more of the **image’s contextual detail, enabling more accurate and meaningful interpretation**.\n![](https://assets.insightmediagroup.io/media/wp-content/uploads/2025/10/1L1YWhun0N6fzCF_aFedKaQ.png)\nApart from higher retrieval accuracy, benefits of ColPali includes:\n 1. **Explainability** : ColPali enables word-level comparison between the query and individual image patches of a document. This allows us to clearly understand and justify why a particular document is deemed more relevant.\n 2. **Reduced Development Effort & Greater Robustness**: By eliminating the need for complex preprocessing pipelines — such as chunking, OCR, and layout parsing — ColPali significantly reduces development time and minimizes potential points of failure.\n 3. **Performance Gains** : Embedding and retrieval processes are faster, resulting in improved overall system responsiveness.\n\n\nNow you know what ColPali is, let’s dive into the code and see if ColPali can resolve the challenges we mentioned earlier!\n## Illustration\nMy code can be found in my [Github](https://github.com/yip-kl?tab=repositories). A few words about it:\n * **Instance** : Running the code requires a machine with A100 GPU\n * **Embedding model** : ColPali has a number of variants, I am using `vidore/colqwen2-v0.1` for demonstration. You can refer to the leaderboard [here](https://huggingface.co/spaces/vidore/vidore-leaderboard) and switch to other models\n * **Agent:** \n**- LLM** : I am using OpenRouter to consume LLM, and the agent is powered by `GPT-4o`. You can substitute it with any multimodal LLM that can take image as input \n- **Orchestration** : `LangGraph` is used to develop the agent\n * **Library used for converting PDF into images** : `pdf2image` which is a wrapper for `poppler`. So please make sure you also have `poppler` installed\n * **Sample data** : “The Impacts and Costs of Climate Change” written by Paul Watkiss et al, which is publicly available [here](https://climate.ec.europa.eu/system/files/2016-11/final_report2_en.pdf)\n\n\n### Results\nThe code is pretty straightforward, so I will jump into conclusion instead: How well can ColPali deal with the problems we have seen in the “Why parsing often falls short?” section?\n**#1. Complex tables cannot be simply be interpreted as texts**\n**Question** : What’s the health impact if temperature rises above pre-industrial level by 3 degrees?\n**Answer** : If the temperature rises above pre-industrial levels by 3 degrees Celsius, the health impact would be significant. `Specifically, it is estimated that by the 2080s, up to 330 million people would be at risk from malaria`. This is a substantial increase compared to the 1920s, where the number at risk was between 50 to 100 million.\nThis is correctly referred from this table in the PDF!\n![](https://assets.insightmediagroup.io/media/wp-content/uploads/2025/10/1u6XGxxl1RidlFh_Fmvnkow.png)\nSource: Page 3 of [The Impacts and Costs of Climate Change](https://climate.ec.europa.eu/system/files/2016-11/final_report2_en.pdf)\n**#2. Disassociation between texts and images**\n**Question** : What’s the modelled cost of climate change, according to the Greenbook declining model?\n**Answer** : According to the Greenbook declining model, the modelled cost of climate change is 7.2 Euro/tCO2\nThe correct answer should be 7.4 Euro/tCO2, but we can see it is close to correct!\n![](https://assets.insightmediagroup.io/media/wp-content/uploads/2025/10/1bkBzLwS5SbIPI9xDNZWGfQ.png)\nSource: Page 46 of [The Impacts and Costs of Climate Change](https://climate.ec.europa.eu/system/files/2016-11/final_report2_en.pdf)\n## Conclusion\nTraditional RAG pipelines struggle with non-textual content. ColPali treats each PDF page as an image, allowing it to process visual layouts, tables, charts, and embedded graphics — formats that standard text parsers often distort or ignore.\nColPali brings vision-language intelligence to RAG, making it far more capable of handling the messy, multimodal reality of enterprise documents.\n_*Unless otherwise noted, all images are by the author_\n## Register Now to Continue Reading\n### Unlock Towards Data Science with a Free Account\nAll articles on Towards Data Science are available to registered users. Create a free account today to access this article and thousands more.\nRegister for freeSign in to your account\n* * *\n![](https://assets.insightmediagroup.io/media/1787657545973_9tepv8.jpeg)\nWritten By\n## [Julian Yip](https://towardsdatascience.com/author/jyipkl/)\n[See all from Julian Yip](https://towardsdatascience.com/author/jyipkl/)\n[Editor's Picks](https://towardsdatascience.com/tag/editors-pick/)[Artificial Intelligence](https://towardsdatascience.com/tag/artificial-intelligence/)[Generative Ai](https://towardsdatascience.com/tag/generative-ai/)[Multimodality](https://towardsdatascience.com/tag/multimodality/)[Rag](https://towardsdatascience.com/tag/rag/)\nShare This Article\n * [](https://www.facebook.com/sharer/sharer.php?u=https%3A%2F%2Ftowardsdatascience.com%2Fbringing-vision-language-intelligence-to-rag-with-colpali%2F&title=Bringing%20Vision-Language%20Intelligence%20to%20RAG%20with%20ColPali)\n * [](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Ftowardsdatascience.com%2Fbringing-vision-language-intelligence-to-rag-with-colpali%2F&title=Bringing%20Vision-Language%20Intelligence%20to%20RAG%20with%20ColPali)\n * [](https://x.com/share?url=https%3A%2F%2Ftowardsdatascience.com%2Fbringing-vision-language-intelligence-to-rag-with-colpali%2F&text=Bringing%20Vision-Language%20Intelligence%20to%20RAG%20with%20ColPali)\n\n\nTowards Data Science is a community publication. Submit your insights to reach our global audience and earn through the TDS Author Payment Program.\n[Write for TDS](https://towardsdatascience.com/submissions/)\n## Related Articles\n * [![](https://assets.insightmediagroup.io/media/wp-content/uploads/2024/03/0QWYON5yhewd6Kqqw-scaled.jpg)](https://towardsdatascience.com/detecting-insecure-code-with-llms-8b8ad923dd98/)\n## [Detecting Insecure Code with LLMs](https://towardsdatascience.com/detecting-insecure-code-with-llms-8b8ad923dd98/)\n[Large Language Models](https://towardsdatascience.com/category/artificial-intelligence/large-language-models/)\nPrompt Experiments for Python Vulnerability Detection\n[Melanie Hart Buehler](https://towardsdatascience.com/author/melanie.h.buehler/)\nMarch 21, 202411 min read\n * [![](https://assets.insightmediagroup.io/media/wp-content/uploads/2023/07/1Nn-77bR4ftcwDbzyLTQ3kw-scaled.jpeg)](https://towardsdatascience.com/running-llama-2-on-cpu-inference-for-document-q-a-3d636037a3d8/)\n## [Running Llama 2 on CPU Inference Locally for Document Q&A](https://towardsdatascience.com/running-llama-2-on-cpu-inference-for-document-q-a-3d636037a3d8/)\n[Large Language Models](https://towardsdatascience.com/category/artificial-intelligence/large-language-models/)\nClearly explained guide for running quantized open-source LLM applications on CPUs using LLama 2, C Transformers, GGML, and LangChain\n[Kenneth Leung](https://towardsdatascience.com/author/kennethleungty/)\nJuly 18, 20234 min read\n * [![](https://assets.insightmediagroup.io/media/wp-content/uploads/2021/10/1TG7z_Kj0MgSGUtehjVaPqQ.jpeg)](https://towardsdatascience.com/benchmarking-time-series-datasets-with-style-2a4e8789a965/)\n## [Benchmarking time series datasets with style](https://towardsdatascience.com/benchmarking-time-series-datasets-with-style-2a4e8789a965/)\n[](https://towardsdatascience.com/bringing-vision-language-intelligence-to-rag-with-colpali/)\nHow to benchmark/spot differences between time series datasets?\n[M Adel](https://towardsdatascience.com/author/muadelm/)\nOctober 20, 20213 min read\n * [![](https://assets.insightmediagroup.io/media/wp-content/uploads/2025/07/FAST_10_wordcloud.png)](https://towardsdatascience.com/topic-model-labelling-with-llms/)\n## [Topic Model Labelling with LLMs](https://towardsdatascience.com/topic-model-labelling-with-llms/)\n[Large Language Models](https://towardsdatascience.com/category/artificial-intelligence/large-language-models/)\nPython tutorial for reproducible labeling of cutting-edge topic models with GPT4-o-mini.\n[Petr Koráb](https://towardsdatascience.com/author/petrkorab/)\nJuly 14, 20256 min read\n * [![](https://assets.insightmediagroup.io/media/wp-content/uploads/2023/09/1Dpe9Lkm6_kyqIlHvphopfQ.png)](https://towardsdatascience.com/highlights-on-large-language-models-at-kdd-2023-fc53440563c3/)\n## [Highlights on Large Language Models at KDD 2023](https://towardsdatascience.com/highlights-on-large-language-models-at-kdd-2023-fc53440563c3/)\n[Deep Learning](https://towardsdatascience.com/category/artificial-intelligence/deep-learning/)\nWere you not able to attend KDD? Learn from my summary about the hottest topic at the conference: LLMs\n[Gabriel Moreira](https://towardsdatascience.com/author/gspmoreira/)\nSeptember 11, 20238 min read\n * [![](https://assets.insightmediagroup.io/media/wp-content/uploads/2019/08/1djuN0_ODnfmyvdsPixcZlg-scaled.jpeg)](https://towardsdatascience.com/heart-disease-prediction-73468d630cfc/)\n## [Heart Disease Prediction](https://towardsdatascience.com/heart-disease-prediction-73468d630cfc/)\n[Artificial Intelligence](https://towardsdatascience.com/category/artificial-intelligence/)\nCleveland Heart Disease(UCI Repository) dataset - classification with various models.\n[Shubhankar Rawat](https://towardsdatascience.com/author/shubhankarrawatsam.1999/)\nAugust 10, 20198 min read\n * [![](https://assets.insightmediagroup.io/media/wp-content/uploads/2024/06/0zUHFxMf1OIYpUXYg-scaled.jpg)](https://towardsdatascience.com/i-invented-a-way-to-speak-to-an-ai-keeping-your-privacy-ddbca5f24e4a/)\n## [I Invented a Way to Speak to an AI, Keeping Your Privacy](https://towardsdatascience.com/i-invented-a-way-to-speak-to-an-ai-keeping-your-privacy-ddbca5f24e4a/)\n[Artificial Intelligence](https://towardsdatascience.com/category/artificial-intelligence/)\nThe tech is called \"Silent Voice.\"\n[Rafe Brena, Ph.D.](https://towardsdatascience.com/author/rafebrena/)\nJune 27, 20248 min read\n * [![](https://assets.insightmediagroup.io/media/wp-content/uploads/2025/09/image-172.jpg)](https://towardsdatascience.com/how-to-enrich-llm-context-to-significantly-enhance-capabilities/)\n## [How to Enrich LLM Context to Significantly Enhance Capabilities](https://towardsdatascience.com/how-to-enrich-llm-context-to-significantly-enhance-capabilities/)\n[Large Language Models](https://towardsdatascience.com/category/artificial-intelligence/large-language-models/)\nLearn how to empower your LLMs by leveraging additional metadata\n[Eivind Kjosbakken](https://towardsdatascience.com/author/oieivind/)\nSeptember 16, 20257 min read\n * [![](https://assets.insightmediagroup.io/media/wp-content/uploads/2026/08/patrick-tomasso-Oaqk7qqNh_c-unsplash-scaled.jpg)](https://towardsdatascience.com/text-watermarking-in-python-catch-whoever-copies-your-writing/)\n## [Text Watermarking in Python: Catch Whoever Copies Your Writing](https://towardsdatascience.com/text-watermarking-in-python-catch-whoever-copies-your-writing/)\n[Large Language Models](https://towardsdatascience.com/category/artificial-intelligence/large-language-models/)\nAI companies quietly watermark billions of words a day. Here’s how to apply the same three families of techniques to your own writing—and what real experiments reveal about which watermarks survive copy-paste, editing, and paraphrasing.\n[Chien Vu Minh](https://towardsdatascience.com/author/chienvu/)\nSeptember 6, 202626 min read\n\n\n * [YouTube](https://www.youtube.com/c/TowardsDataScience)\n * [X](https://x.com/TDataScience)\n * [LinkedIn](https://www.linkedin.com/company/towards-data-science/)\n * [Threads](https://www.threads.net/@towardsdatascience)\n * [Bluesky](https://bsky.app/profile/towardsdatascience.com)\n\n\n[](https://towardsdatascience.com/)\nYour home for data science and AI. The world's leading publication for data science, data analytics, data engineering, machine learning, and artificial intelligence professionals.\nSubscribe to Our Newsletter[Write For TDS](https://towardsdatascience.com/submissions/)•[About](https://towardsdatascience.com/about-towards-data-science-d691af11cc2f/)•[Advertise](https://contact.towardsdatascience.com/advertise-with-towards-data-science)•[Privacy Policy](https://towardsdatascience.com/privacy-policy/)•[Terms of Use](https://towardsdatascience.com/website-terms-of-use/)\n© Insight Media Group, LLC 2026•[The New Stack](https://thenewstack.io)•[roadmap.sh](https://roadmap.sh)•[Creator Network](https://insightmediagroup.io/creator-network)\n#### Recommended Articles \nClose\n * [ ![](https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/607686/image?etag=1782161464448&width=500&height=500) How Relevance Models Foreshadowed Transformers for NLP ](https://towardsdatascience.com/how-relevance-models-foreshadowed-transformers-for-nlp/ \"How Relevance Models Foreshadowed Transformers for NLP\")\n * [ ![](https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/607639/image?etag=1781250125848&width=500&height=500) Javascript Fatigue: HTMX Is All You Need to Build ChatGPT — Part 1 ](https://towardsdatascience.com/javascript-fatigue-you-dont-need-js-to-build-chatgpt/ \"Javascript Fatigue: HTMX Is All You Need to Build ChatGPT — Part 1\")\n * [ ![](https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/607930/image?etag=1782124566054&width=500&height=500) The Subset Sum Problem Solved in Linear Time for Dense Enough Inputs ](https://towardsdatascience.com/subset-sum-problem-solved-in-linear-time-for-dense-enough-inputs/ \"The Subset Sum Problem Solved in Linear Time for Dense Enough Inputs\")\n * [ ![](https://h030.towardsdatascience.com/rest/v2/contentStores/d7d3a4ab-7caf-4383-bbd2-fbf7fb4f1787/items/3660/image?etag=1780176246070&width=500&height=500) Does Your Company Have a Data Strategy? ](https://towardsdatascience.com/does-your-company-have-a-data-strategy-e8944dbec4ae/ \"Does Your Company Have a Data Strategy?\")\n\n\n", "body_base_path": "attachments/191976f37240/pages/", "body_markdown_path": "attachments/191976f37240/pages/archived-body.md", "local_html": "attachments/191976f37240/pages/archived-body.html", "body_source": "prior_capture" }