Arcfield’s Post

There's no better way to end my internship at 🐢 Giskard than by publishing an article! In this article, I investigate the optimal number of Q&A required for fine-tuning an LLM in order to enhance a RAG system. I also study the impact of fine-tuning with distracting context on model performance. The goal is to identify what constitues an efficient and cost-effective strategy for fine-tuning in that context. Link: https://lnkd.in/d3WUmSrc Key Findings: - Adding even a single distracting document to the context when training an LLM helps it distinguish distracting context from supporting facts during inference, thus making it more efficient. This is especially interesting in a RAG setting, where the retrieval step might bring out distracting context; - Fine-tuning Mistral-7B-Instruct for a single epoch on under 1000 training samples is very quick and can significantly improves results. I would like to thank the entire team and especially Rabah Khalek, Benoît Malézieux, Matteo Dora, Pierre Le Jeune and Luca Martial for their guidance during those past 6 months. I learnt so much alongside the team during the various projects I participated in, especially about multimodal models, RAGs and fine-tuning. I am extremely grateful to you guys!

In the latest issue of the Urban Technology at University of Michigan, program director Bryan Boyer revisits a core hypothesis of the degree program—more data, more feedback loops, better buildings and cities—and we hear from a few students about their summer internships, who are themselves creating new feedback loops into our curriculum. Read issue 164 of the Urban Technology at University of Michigan newsletter: https://lnkd.in/gcuAXQbi

I’m excited to begin a summer internship with Open Corridor. Over the next couple of months, I’ll be diving into the world of applied mathematics and combining it with my love of coding, all while contributing to projects that have a positive impact on our planet. Environmental sustainability is one of the most important challenges to date, requiring collaborations from governments, organisations and individuals alike - I am privileged to play a small part in this effort! #environment #environmentalsustainability #appliedmathematics #datascience #pythonprogramming

Salvador Tranquilino–Ramos, a May graduate with a bachelor’s degree in data science with a 4.0 GPA, spent his time with UIC completing internships, undergraduate research, and student teaching. This prolific student is now starting a career with the online reservation platform Tock, and is also starting an Online Master of Computer Science in Data Science degree with UIUC. Learn more about Tranquilino–Ramos here: https://shorturl.at/ZptKE #UIC #UICComputerScience #UICCollegeofEngineering

Let's gooo, Nicole! 🚌🚗🚊 Nicole Stoddard, public policy '25, completed a summer internship with the U.S. Department of Transportation's Federal Motor Carrier Safety Administration. Stoddard served as a Data Analyst intern, studying crash data across Massachusetts, New Hampshire, Vermont, and Maine. In this role, she identified the greatest risk factors for accidents and reported information back to state offices to address prevention. Want to learn more about Nicole's experience in this internship? Read more at www.umass.edu/spp! #umass #umassspp #publicpolicy #transitpolicy #goumass

Want an inside look into our summer internship program? On this week's podcast episode, we're joined by one of our interns, Jonah Nichols, who is working within our Data and Analytics division as a data engineering intern. He shares about himself, his experience working with Moser, and the parallels between his work for Moser and his coursework as a computer engineering student at Purdue University. Listen to Jonah's story: https://buff.ly/3OQfG7y #TechIntern #Honeycomb #MachineLearning #DataEngineering

🦈 Excited to share my recent Shark Attacks Data Analysis Project! 🌊📊 I collected, cleaned, and analyzed a dataset that contains shark 🦈 attack reports spanning over 100 years, starting from 1900. 📈 Understanding shark attack patterns is crucial for enhancing public safety and guiding conservation efforts. Accurate data aids stakeholders in identifying trends, assessing risk factors, and crafting effective prevention strategies. 🦈🔍 Check out the video below for a detailed breakdown of my analysis. Special thanks to Quantum Analytics NG for providing the dataset during my internship. 🙏🏼 #DataAnalysis #SharkAttacks #PublicSafety #ConservationEfforts #QuantumAnalytics

Happy to share that I've successfully wrapped up the third task of my internship at SYNC INTERN'S —the Boston House Prediction project. 🏠 Step 1: Data Exploration and Preprocessing Explored and preprocessed the Boston Housing dataset, gaining valuable insights into the features and their impact on housing prices. 🔍 Step 2: Feature Selection and Interpretation Selected key features based on their strong correlation with the target variable. Crafted meaningful interpretations of these features, enhancing our understanding of the factors influencing housing prices. 🤖 Step 3: Model Selection Implemented various regression models, including Linear Regression, Decision Tree, and Random Forest, to predict median home values. Conducted a thorough analysis to determine the most effective model. The Project Link : https://lnkd.in/dNzHSSTi Grateful for the incredible learning experience and the opportunity to contribute to impactful projects. #Internship #DataScience #MachineLearning #BostonHousePrediction #Achievements

Businesses have found their next employee by utilizing the Internship program with offer with Emporia State University. Learn more in today's Tuesday Talk!

"Successfully completed a one-month data science internship, diving deep into the world of tech and data. Excited to apply my newfound skills and knowledge in the next chapter of my journey! #DataScience #technohacksinternship "