The hottest Data science Substack posts right now

And their main takeaways
Category
Top Technology Topics

The Tech Buffet #2: How To Use LangChain to Perform Question Answering Over Documents

The Tech Buffet • 59 implied HN points • 06 Sep 23
🕹 Technology Data science
  1. You can use LangChain to build a question-answering system that works with documents. It helps you fetch answers from documents effortlessly.
  2. The process involves loading a document, splitting it into manageable chunks, and then using these chunks to find answers. This way, you have context to support the answers generated.
  3. It's important to keep experimenting and refining your system for better answers. Check out more details in the LangChain documentation for tips and improvements.

KCOR v6 enables radical transparency into vaccine data

Steve Kirsch's newsletter • 4 implied HN points • 22 Dec 25
🔬 Science Data science
  1. KCOR v6 fits a Gompertz gamma‑frailty model to cohorts' cumulative hazards to remove heterogeneity and allow fair comparisons between vaccinated and unvaccinated groups.
  2. Applied to Czech data, KCOR shows a net harm signal (KCOR > 1) for mRNA COVID vaccines over time, with boosters appearing especially harmful in the weeks after vaccination.
  3. The method depends on assumptions (Gompertz mortality, gamma frailty, and that vaccine harm subsided by mid‑2022) and has limits: it can miss very early post‑shot spikes, long‑term monotonic risk increases, and non‑proportional hazard effects.

Prompt Engineering Lecture

Elvis's Blog • 58 implied HN points • 19 Feb 23
🕹 Technology Data science
  1. The post is about a special lecture on prompt engineering techniques for language models.
  2. The lecture is divided into four parts: Introduction to Prompt Engineering, Advanced Techniques, Tools and Applications, and Conclusion with Future Directions.
  3. The post provides links to notebook, slides, and GitHub for further exploration.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

ChatGPT and Supply Chain: Podcast Guest, Three Lessons, and Research Opportunities

Mike Talks AI • 58 implied HN points • 13 Jun 23
🕹 Technology Data science
  1. Supply chain professionals can use ChatGPT as a 'loss leader' to educate leaders about AI's potential for supply chains.
  2. ChatGPT can help supply chain teams build more AI algorithms by breaking down syntax barriers and expanding team capabilities.
  3. Exploring how ChatGPT can turn vast supply chain data into valuable insights is an important research opportunity.

Edge 439: SSMs with Attention, Understanding Zamba

TheSequence • 112 implied HN points • 15 Oct 24
🕹 Technology Data science
  1. Combining state space models (SSMs) with attention layers can create better hybrid architectures. This fusion allows for improved learning capabilities and efficiency.
  2. Zamba is an innovative model that enhances learning by using a mix of Mamba blocks and a shared attention layer. This approach helps it manage long-range dependencies more effectively.
  3. The new architecture reduces the computational load during training and inference compared to traditional transformers, making it more efficient for AI tasks.

DRAGIN: Dynamic RAG Based On Real-Time Information Needs Of LLMs

Cobus Greyling on LLMs, NLU, NLP, chatbots & voicebots • 19 implied HN points • 26 Mar 24
🕹 Technology Data science
  1. Dynamic Retrieval Augmented Generation (RAG) improves the way information is retrieved and used in large language models during text generation. It focuses on knowing exactly when and what to look up.
  2. Traditional RAG methods often use fixed rules and may only look at the most recent parts of a conversation. This can lead to missed information and unnecessary searches.
  3. The new framework called DRAGIN aims to make data retrieval smarter and faster without needing further training of the language models, making it easy to use.

The Sequence Chat: Thinking About Transformers as Computers

TheSequence • 105 implied HN points • 30 Oct 24
🕹 Technology Data science
  1. Transformers are changing AI, especially in how we understand and use language. They're not just tools; they act more like computers in some ways.
  2. The way transformers can adapt and scale is really impressive. It's like they can learn and adjust in ways traditional computers can't.
  3. Thinking of transformers as computers opens up new ideas about how we approach AI. This perspective can help us find new applications and improve our understanding of tech.

The Sequence Radar #472: Remember this Name: Ndea

TheSequence • 77 implied HN points • 19 Jan 25
🕹 Technology Data science
  1. Ndea is a new AI lab aiming to create artificial general intelligence (AGI) with a unique approach called guided program synthesis. This approach allows models to learn efficiently from fewer examples.
  2. Francois Chollet, a well-known AI expert, is leading Ndea. He believes current deep learning methods have limitations and wants to explore new ideas for better AI development.
  3. The goal of Ndea is to drive quick scientific advancements by combining program synthesis with deep learning, aiming to tackle tough challenges and possibly discover new scientific frontiers.

April Progress Prizes / Updates

Vesuvius Challenge • 38 implied HN points • 23 May 25
🕹 Technology Data science
  1. New techniques for analyzing scroll shapes are improving the way we handle and segment data. This means we can understand and work with historical documents much better.
  2. There have been exciting updates in scroll deformation methods, which can help in restoring the original shapes of ancient scrolls. This makes analyzing them easier and more accurate.
  3. The new developments in fiber analysis provide important information that can help reconstruct ancient writing surfaces. This can lead to better ways to unroll and study papyrus materials.

Prompt-RAG

Cobus Greyling on LLMs, NLU, NLP, chatbots & voicebots • 19 implied HN points • 20 Mar 24
🕹 Technology Data science
  1. Prompt-RAG is a new method that improves language models without using complex vector embeddings. It simplifies how we retrieve information to answer questions.
  2. The process involves creating a Table of Contents from documents, selecting relevant headings, and generating responses by injecting context into prompts. It makes handling data easier.
  3. While this method is great for smaller projects and specific needs, it still requires careful planning when constructing the documents and managing costs related to token usage.

What is Generative AI?

Technically • 27 implied HN points • 22 Jul 25
🕹 Technology Data science
  1. Generative AI predicts not just numbers or yes/no answers but creates full sentences, images, and even videos from prompts.
  2. There are various types of Generative AI models, with the main ones being Transformers for text and Diffusion models for images.
  3. Despite its advancements, Generative AI is still rooted in the basic principles of machine learning, which involves learning patterns from data.

Performing Multiple LLM Calls & Voting On The Best Result Are Subject To Scaling Laws

Cobus Greyling on LLMs, NLU, NLP, chatbots & voicebots • 19 implied HN points • 19 Mar 24
🕹 Technology Data science
  1. Making more calls to Large Language Models (LLMs) can help with simple questions but may actually make it harder to answer tough ones.
  2. Finding the right number of calls to use is crucial for getting the best results from LLMs in different tasks.
  3. It's important to design AI systems carefully, as just increasing the number of calls doesn't always mean better performance.

Agent Evaluation Playbook

Pratik’s Pakodas 🍿 • 27 implied HN points • 08 Jul 25
🕹 Technology Data science
  1. To make good AI agents, it's important to have a solid evaluation process. This can help ensure they're performing well in real-world situations.
  2. Creating a system that tracks and measures the agents' performance can lead to better results. Like building a pipeline that continuously tests and improves agents.
  3. Using a leaderboard to compare agents based on performance, cost, and speed can help guide improvements and make smarter decisions.

dbt Labs acquires Transform - who's next?

Data Thoughts • 119 implied HN points • 19 Feb 23
💼 Business Data science
  1. dbt Labs has bought Transform, and more companies in the data field might be sold or closed soon. This could lead to big changes in the industry.
  2. Data teams are seen as a 2nd order need for businesses, meaning they aren't absolutely necessary. Companies may cut these teams first when they need to save money.
  3. To get the best value from tools, data practitioners should focus on essential needs rather than extra features. This means keeping an eye on what really matters in the data ecosystem.

Artificial intelligence and "big data" cannot replace public opinion polls | #213 - April 10, 2023

G. Elliott Morris's Newsletter • 119 implied HN points • 10 Apr 23
🇺🇸 U.S. Politics Data science
  1. Artificial intelligence and big data cannot fully replace public opinion polls, as they rely on polls for calibration and may not be as reliable for all groups.
  2. Changes in polling methods, like switching from phone to online surveys, can impact results, highlighting the importance of consistency over time.
  3. Studies show genuine change in attitudes, like increasing racial liberalism, but also caution against biases affecting survey responses.

World Models are Coming and They are Awesome

TheSequence • 84 implied HN points • 08 Dec 24
🕹 Technology Data science
  1. This week saw the release of two exciting world models that can create 3D environments from simple prompts. These models are important for advancing AI's abilities in various fields.
  2. DeepMind's Genie 2 can generate interactive 3D worlds and simulate realistic object interactions, making it very useful for AI training and game development.
  3. World Labs has introduced a user-friendly system for designing 3D spaces, allowing artists to create and manipulate environments easily, which can help in game prototyping and creative workflows.

AIM Daily XO

Sector 6 | The Newsletter of AIM • 39 implied HN points • 05 Dec 23
🚌 Education Data science
  1. AIM has been ranking graduate programs for eight years, focusing on Data Science programs in India for 2023. They use surveys and research to create these rankings.
  2. This year's rankings include both on-campus and online/hybrid postgraduate programs. This helps students find options that fit their learning style.
  3. A strong program is one that scores well across various areas, showing its quality and value to students.

Edge 459: Quantization Plus Distillation

TheSequence • 77 implied HN points • 24 Dec 24
🕹 Technology Data science
  1. Quantized distillation helps make deep neural networks smaller and faster by combining two techniques: knowledge distillation and quantization.
  2. This method transfers knowledge from a high-precision model (teacher) to a low-precision model (student) without losing much accuracy.
  3. Using soft targets from the teacher model can reduce problems that often come with using simpler models, keeping performance strong.

LLMs Training SLMs

Cobus Greyling on LLMs, NLU, NLP, chatbots & voicebots • 19 implied HN points • 12 Mar 24
🕹 Technology Data science
  1. Orca-2 is designed to be a small language model that can think and reason by breaking down problems step-by-step. This makes it easier to understand and explain its thought process.
  2. The training data for Orca-2 is created by a larger language model, focusing on specific strategies for different tasks. This helps the model learn to choose the best approach for various challenges.
  3. A technique called Prompt Erasure helps Orca-2 not just mimic larger models but also develop its own reasoning strategies. This way, it learns to think cautiously without relying on direct instructions.

How RLHF actually works

Democratizing Automation • 306 implied HN points • 21 Jun 23
🕹 Technology Data science
  1. RLHF works when there is a signal that vanilla supervised learning alone doesn't work, like pairwise preference data.
  2. Having a capable base model is crucial for successful RLHF implementation, as imitating models or using imperfect datasets can greatly affect performance.
  3. Preferences play a key role in the RLHF process, and collecting preference data for harmful prompts is essential for model optimization.

Edge 457: Can we Distill Specific Knowledge in LLMs? An Intro to Attention-Based Distillation

TheSequence • 77 implied HN points • 17 Dec 24
🕹 Technology Data science
  1. Attention-based distillation (ABD) is a method that helps smaller models learn from larger models by mimicking their attention patterns. This can make the smaller models perform better with fewer resources.
  2. Unlike traditional methods that just look at output predictions, ABD focuses on the reasoning process of the larger model. This leads to a deeper understanding and better results for the smaller model.
  3. Using ABD can produce student models that perform well even when they have less complexity. This is useful for applications where efficiency is key.

The Sequence Research #466: Small but Migthy, Diving Into Microsoft Phi-4

TheSequence • 70 implied HN points • 10 Jan 25
🕹 Technology Data science
  1. Microsoft's Phi-4 is a new language model that's smaller in size but powerful in performance. It shows that high-quality data can make a big difference in AI.
  2. Phi-4 has 14 billion parameters, which means it can handle complex language tasks effectively. This model builds on the success of earlier Phi models.
  3. The innovations in Phi-4 come from its unique approach to training, focusing on pre-training, mid-training, and post-training stages to enhance its capabilities.

Run Large Language Model On Your Own Computer

The Beep • 19 implied HN points • 10 Mar 24
🕹 Technology Data science
  1. You can run large language models, like Llama2, on your own computer using a tool called Ollama. This allows you to use powerful AI without needing super high-tech hardware.
  2. Setting up Ollama is simple. You just need to download it and run a couple of commands in your terminal to get started.
  3. Once it's running, you can interact with the model like you would with any chatbot. This means you can type prompts and get responses directly from your own machine.

Not All Layers Are Equal

Gonzo ML • 63 implied HN points • 31 Jan 25
🕹 Technology Data science
  1. Not every layer in a neural network is equally important. Some layers play a bigger role in getting the right results, while others have less impact.
  2. Studying how information travels through different layers can reveal interesting patterns. It turns out layers often work together to make sense of data, rather than just acting alone.
  3. Using methods like mechanistic interpretability can help us understand neural networks better. By looking closely at what's happening inside the model, we can learn which parts are doing what.

Transformer^2: Self-adaptive LLMs

Gonzo ML • 63 implied HN points • 27 Jan 25
🕹 Technology Data science
  1. Transformer^2 uses a new method for adapting language models that makes it simpler and more efficient than fine-tuning. Instead of retraining the whole model, it adjusts specific parts, which saves time and resources.
  2. The approach breaks down weight matrices through a process called Singular Value Decomposition (SVD), allowing the model to identify and enhance its existing strengths for various tasks.
  3. At test time, Transformer^2 can adapt to new tasks in two passes, first assessing the situation and then applying the best adjustments. This method shows improvements over existing techniques like LoRA in both performance and parameter efficiency.

Microsoft’s New Love

Sector 6 | The Newsletter of AIM • 39 implied HN points • 17 Nov 23
🕹 Technology Data science
  1. Large language models (LLMs) like ChatGPT are powerful but costly to run and customize. They require a lot of resources and can be tricky to adapt for specific tasks.
  2. Small language models (SLMs) are emerging as a better option because they are cheaper to train and can give more accurate results. They also don't need heavy hardware to operate.
  3. Many companies are starting to focus on developing small language models due to their efficiency and effectiveness, marking a shift in the industry.

Self-Reflective Retrieval-Augmented Generation (SELF-RAG)

Cobus Greyling on LLMs, NLU, NLP, chatbots & voicebots • 19 implied HN points • 04 Mar 24
🕹 Technology Data science
  1. SELF-RAG is designed to improve the quality and accuracy of responses from generative AI by allowing the AI to reflect on its own outputs and decide if it needs to retrieve additional information.
  2. The process involves generating special tokens that help the AI evaluate its answers and determine whether to get more information or stick with its original response.
  3. Balancing efficiency and accuracy is crucial; too much focus on speed can lead to wrong answers, while aiming for perfect accuracy can slow down the system.

Visualizing the Chain Rule

The Palindrome • 4 implied HN points • 22 Dec 25
🕹 Technology Data science
  1. The chain rule is essential in machine learning because it lets you compute gradients of composite functions, which you need for gradient descent and fitting models.
  2. The single-variable rule is simple, but with many parameters you must handle vector-valued functions and the math gets more complicated in the multivariable case.
  3. Each parameter's gradient is a sum over model outputs: the loss's sensitivity to each output times that output's sensitivity to the parameter, which is equivalent to multiplying gradients/Jacobians to propagate derivatives.

Do Not Spend too Much Time "Getting Good" at Dealing with Current AML GPT LLMs

Brad DeLong's Grasping Reality • 169 implied HN points • 04 Mar 24
🕹 Technology Data science
  1. It's uncertain how current AML GPT LLMs will be most useful in the future, so spending too much time trying to master them may not be the best approach.
  2. Proper prompting is crucial when working with AML GPT LLMs as they can be capable of more than initially apparent. Good prompts can make tasks that seem impossible into achievable ones.
  3. Understanding the thought processes and effective way to prompt AML GPT LLMs is essential, as their responses can vary based on subtle changes or inadequate prompting.

Everybody Likes RAG

Sector 6 | The Newsletter of AIM • 39 implied HN points • 15 Nov 23
🕹 Technology Data science
  1. RAG stands for retrieval-augmented generation, which is becoming really popular in the tech world. People are eager to use it for their work.
  2. It offers many benefits like better access to current information and helps to verify sources. It's also efficient and cost-effective.
  3. Some see RAG as just a fancy version of prompt engineering, but others think it's essential for growing business applications.

How to destroy a world

Data People Etc. • 53 implied HN points • 24 Feb 25
🕹 Technology Data science
  1. Frameworks can be used for both building and breaking worlds. It's important to understand how to exploit weaknesses in these structures.
  2. To weaken a dominant system, you can undermine its narrative, disrupt key players, and challenge established norms. This approach can create doubts and resistance.
  3. Destroying a world can teach us about resilience. Strengthening systems and protocols is crucial to support and maintain their relevance in changing times.

Scale, Scale, Scale

Gradient Flow • 179 implied HN points • 05 May 22
🕹 Technology Data science
  1. The importance of scale in AI startups highlighted by the proficiency in distributed systems over ML and AI.
  2. Exploring the impact of distributed computing on machine learning and AI through metrics.
  3. Insights from the Data Exchange podcast on topics like scaling language models, applying ML to optimization, and blending data science with domain expertise.

The Tech Buffet #13: Getting a RAG To Work Well Is Hard - 5 Blog Posts To Become a RAG Master

The Tech Buffet • 39 implied HN points • 13 Nov 23
🕹 Technology Data science
  1. RAG systems have limitations, like difficulties in effectively retrieving complex information from text. It's vital to understand these limits to use RAGs successfully.
  2. Improving RAG performance involves strategies like cleaning your data and adjusting chunk sizes. These tweaks can help make RAG systems work a lot better.
  3. RAGs may not meet all needs in specialized fields, like insurance, since they sometimes miss important details in lengthy documents. Other methods might be needed for these complex queries.

Robotics is Inching Towards it ChatGPT Moment

TheSequence • 84 implied HN points • 03 Nov 24
🕹 Technology Data science
  1. Robots are getting smarter with new tech, especially using large language models, which help them learn and do tasks better.
  2. MIT's new technique helps robots understand different types of data, making them more capable and efficient in their work.
  3. There’s a big push for robots to interact more naturally with humans, like being able to feel and handle objects carefully, which can improve everyday tasks.

The Sequence Chat: Why are Foundation Models so Hard to Explain and What are we Doing About it?

TheSequence • 77 implied HN points • 27 Nov 24
🕹 Technology Data science
  1. Foundation models are really complex and hard to understand. They act like black boxes, which makes it tough to know how they make decisions.
  2. Unlike older machine learning models, these large models have much more advanced capabilities but also come with bigger interpretability challenges.
  3. New fields like mechanistic interpretability and behavioral probing are trying to help us figure out how these complex models work.