IBM’s Post

IBM Whatever data solution it may be, extracting meaningful information from unstructured text, especially multilingual unstructured text, needs multilingual metadata. Is there any data solution, AI or NOT, that can answer the following questions of business intelligence? "Who, in the Ontario province of Canada, have new US patents granted on the nearest Tuesday (Eastern Time), when the USPTO releases the newly granted US patents on a weekly basis?" "Who, in the "江蘇" province of China, have new US patents granted on the nearest Tuesday (Eastern Time), when the USPTO releases the newly granted US patents on a weekly basis?" With our intellectual property (IP), a Chinese-English multilingual metadata, we can get the full list answers for the above questions. This is a fact. Do you or any of your contacts need our expertise/IP to do the data analysis that AI can't do? Metadata is an enabler. It is like a treasure map for treasure hunting. Without metadata, like a treasure map, NO data can be found/retrieved, even by the most advanced technologies, like AI, high-end chips, supercomputers, etc. https://lnkd.in/g-aJFnXR

IBM Granite model family…

In Europe, influencers need to label posts as advertisements when sharing on social media. Now, I wouldn't call myself an influencer (far from it! 😄), but since I work for IBM, I suppose there's a bit of bias here. We haven’t done the best job promoting our Granite family of GenAI models until now, so I was pretty excited to see this post this week. 🙌 I’ve been working with the team behind 𝐆𝐫𝐚𝐧𝐢𝐭𝐞 𝐟𝐨𝐫 𝐂𝐨𝐝𝐞 on Watsonx Code Assistant, while 𝐆𝐫𝐚𝐧𝐢𝐭𝐞 𝐟𝐨𝐫 𝐋𝐚𝐧𝐠𝐮𝐚𝐠𝐞 has been a lifesaver in my user research—especially when it comes to speeding up tasks. Who remembers the days of manually summarizing meetings after calls? Or when 20 user interviews meant weeks of transcriptions? Those days are long gone, and I definitely don’t miss them! It’s amazing how AI is quietly transforming the way we work, step by step. That said, when we share things on LinkedIn, aren’t we all kind of influencers in our own circles? I don't know because I'm not an influencer—as said—but just something to think about! 😉 Check out watsonx today 😜 (Advertisement (Just to be safe 😆 )) Karen-ellen Browne, Alvin Tan, Angel Montesdeoca, Christine Yu, Katharina Schippert, Corrie Kwan, Zach Nilsson, John H. Bailey, Ph.D., Florian Niethammer, Ana Manrique, Carlos Rosemberg, Ellen Kolstø, Ashley Profozich #AI #GenAI #UserResearch #watsonx #LLM #Code #AIinDesign #IBMAI

Explore the 4 Granite #llm specialized modalities for coding, geo-spatial, language and time series optimization. Which one should you use for to improve your next project?

🚀 In this article, I explore the differences and similarities between traditional text similarity algorithms and Large Language Models (LLMs). By carefully selecting the appropriate method based on the specific use case, we can leverage these powerful tools to enhance our text analysis and achieve more accurate and meaningful results. 🌟 #TextAnalytics #SimilarityAlgorithms #LLMs #DataScience #IBM #AIEngineering

Want to demystify large language models (LLMs) without delving into the intricacies of math or coding? Here's a simplified explanation of this complex technical subject. #LanguageModels #Technology #SimplifiedExplanation

AI will not replace programming. It only changes the way of programming. For example- Traditional programming: 1. Define a function. 2. write code to call function 3. execute. 4. print the value AI programming: 1. Define a function. 2. Write prompt template to describe a function in natural language. 3. Invoke LLM. 4. Respond in natural language. THERE WERE SYSTEM PROGRAMS TO DO BACKGROUND PROCESSING, NOW EXTRA LAYER OF LLM TO HUMANIZE THE PROGRAMMING FRAMEWORK. #LLM , #GENAI #PRODUCTS

I am thrilled to announce the completion of my Project of #Python_Programming , #Task_3 as assigned by #CodeAlpha . In this Task_3 Create a text-based chatbot that can have conversations with users. You can use natural language processing libraries like NLTK or spaCy to make your chatbot more conversational. For those interested in the technical details, you can find the full project repository on #GitHub: https://lnkd.in/g3CK3hxF #ProjectCompletion #PythonProgramming CodeAlpha #GitHub #TechCommunity #InteractiveDesign #SkillDevelopment #TaskCompletion #CodeAlpha #TechCommunity #LearningJourney

Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking This paper (Zelikman et al.) presents a generalization of the earlier work on Self-Taught Reasoner (STaR, Zelikman et al. 2022) in which LMs learn to generate rationales at each token to explain future text, improving their predictions. Quiet-STaR addresses key challenges of the earlier approach, including 1) the computational cost of generating continuations, 2) the fact that the LM does not initially know how to generate or use internal thoughts, and 3) the need to predict beyond individual next tokens. Major points: - Quiet-STaR enhances generative language models by enabling them to generate internal rationales before making text predictions, building on the Self-Taught Reasoner concept. - Quiet-STaR uses a process involving three core activities: think (generate rationale candidates), talk (assess utility of rationales), and learn (optimize rationale generation). - Applying Quiet-STaR to models based on the Mistral architecture has shown significant improvements in reasoning tasks like GSM8K and CommonsenseQA WITHOUT task-specific finetuning. https://lnkd.in/gyrqj9qj #LLMs #InternalModels #NLP

7 ways to integrate AI tools into your software development workflow: 1. Code Assistance and Completion 2. Automated Testing and Debugging 3. Natural Language Processing for Code Generation 4. Enhanced Search and Knowledge Management 5. Project Management and Collaboration 6. Continuous Integration and Deployment (CI/CD) 7. Documentation and Knowledge Sharing Integrate these AI tools to enhance efficiency, reduce errors, and focus more on innovation. #tipsforsoftdevs #vionso #aitools