The hottest Research Substack posts right now

And their main takeaways
Category
Top Health Politics Topics

Tuesday Links

Niko McCarty • 19 implied HN points • 28 May 24
🔬 Science Research
  1. New AI models are being developed to design DNA sequences and create new antibodies. These advancements could help in understanding genetic activities better.
  2. Researchers have found a clever way to help microbes absorb molecules that usually cannot enter cells by attaching them to compounds that can. This could lead to new methods in biotechnology.
  3. Joining cancer trials might not help patients live longer, according to a study. It's important for patients to consider this when thinking about trial participation.

The Future Is Already Here, It’s Just Not Evenly Distributed

Am I Stronger Yet? • 172 implied HN points • 16 Dec 24
🕹 Technology Research
  1. AI tools have amazing strengths but can also be really weak in some areas. This makes their effectiveness uneven, depending on what task you're trying to do.
  2. People often aren't using AI tools to their full potential. Many are not even trying them out, which means they miss out on big opportunities.
  3. To get the most from AI, you need to be creative and put effort into how you use it. A great prompt can lead to big breakthroughs, while a simple request might not yield good results.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

C57BL/6 (Beta)

Niko McCarty • 19 implied HN points • 25 May 24
🔬 Science Research
  1. In 2032, scientists created computer emulations of mice, including their entire anatomy and brain. This was only possible for a few organizations with strong computing power.
  2. The military used these emulators to test how drugs could enhance mouse performance, but some results were secretly tested on prisoners, raising ethical concerns.
  3. The NIH gave access to emulators mainly to select academic institutions, leading to a flood of biomedical papers. This made their findings influential in clinical trials, affecting millions of people.

‘Lab leak’ in the biological warfare century – clear and present danger

Heterodox STEM • 56 implied HN points • 13 Jul 25
🔬 Science Research
  1. The idea that COVID-19 came from a lab leak has been heavily debated, but there's strong evidence suggesting it originated naturally. Many scientists have already concluded that the virus did not come from the Wuhan Institute of Virology.
  2. Misinformation about vaccines and the origins of COVID-19 can create distrust in science, making it harder for public health efforts to succeed. It's essential to rely on accurate information to manage potential biological threats.
  3. The rapid development and deployment of mRNA vaccines are crucial for defending against future biological attacks. These vaccines can be produced quickly and efficiently, which is vital for protecting public health.

AI Roundup 129: Personal Superintelligence

Artificial Ignorance • 50 implied HN points • 01 Aug 25
🕹 Technology Research
  1. Meta is working on a personal superintelligence for everyone, focusing on AI that understands people deeply and helps them achieve their goals.
  2. Builder.ai, a company that promised to revolutionize app development with AI, has gone bankrupt after fraud accusations, highlighting issues in the startup world with misleading AI claims.
  3. China is positioning itself as a leader in open-source AI development, looking to counter U.S. restrictions while boosting its own AI industry through significant state investment.

New applied metascience lab - Future House announced + some thoughts on UK applied metascience

James W. Phillips' Newsletter • 98 implied HN points • 01 Nov 23
🕹 Technology Research
  1. A new applied metascience lab called Future House has been announced, with a focus on using AI to accelerate scientific research.
  2. Future House aims to create an 'AI scientist' that can independently develop hypotheses by analyzing scientific papers.
  3. The UK needs to prioritize initiatives like Future House to orient to new opportunities and empower quality talent in research.

RLHF learning resources in 2024

Democratizing Automation • 435 implied HN points • 12 Jan 24
🕹 Technology Research
  1. The post shares a categorized list of resources for learning about Reinforcement Learning from Human Feedback (RLHF) in 2024.
  2. The resources include videos, research talks, code, models, datasets, evaluations, blog posts, and other related materials.
  3. The aim is to provide a variety of learning tools for individuals with different learning styles interested in going deeper into RLHF.

Looking for your keys under the light

Trevor Klee’s Newsletter • 671 implied HN points • 13 Jun 23
🔬 Science Research
  1. When searching for something, we tend to look where it is easiest to see, even if it might not be the best place to find it.
  2. This behavior can lead to wasting time and effort on ineffective or inefficient search strategies.
  3. It is important to be mindful of not getting stuck looking in familiar or visible places, but to explore all possibilities.

On Incomprehensible Mystery

The Heart Attack Diet • 79 implied HN points • 06 Dec 23
🏥 Health & Wellness Research
  1. Understanding complex mysteries often involves asking the right questions and breaking down the problem into simpler elements. Once the core questions are identified, solutions become clearer.
  2. History shows that even the most intricate scientific challenges eventually yield to persistent inquiry and the pursuit of knowledge. What may seem incomprehensible at first can become simple with the right approach.
  3. Science is not just about conducting studies and publishing results, but about finding answers through experimentation and continual questioning. The key lies in identifying the right questions and trusting in replicable, well-designed studies.

Crazy Science Idea: Tell People Something Is Possible When You Actually Have No Clue

Atlas of Wonders and Monsters • 593 implied HN points • 03 Aug 23
🔬 Science Research
  1. Sometimes telling people something is possible, even if you're unsure, can lead to faster progress in finding solutions.
  2. Encouraging the pursuit of crazy ideas, even if they may not be true, can spark innovation and breakthroughs.
  3. Distorting facts slightly to make crazy ideas seem less crazy could potentially inspire more discovery and creativity.

The Sequence Research #558: The New Reinforcement Learning from Internal Feedback Allows LLMs to Reason Without External Rewards

TheSequence • 70 implied HN points • 06 Jun 25
🕹 Technology Research
  1. Reinforcement learning is a key way to help large language models think and solve problems better. It helps models learn to align with what people want and improve accuracy.
  2. Traditional methods like RLHF require a lot of human input and can be slow and costly. This limits how quickly models can learn and grow.
  3. A new approach called Reinforcement Learning from Internal Feedback lets models learn on their own using their own internal signals, making the learning process faster and less reliant on outside help.

What the executive order means for openness in AI

AI Snake Oil • 489 implied HN points • 31 Oct 23
🕹 Technology Research
  1. The executive order on AI strives to address various benefits and risks, impacting openness in the AI landscape.
  2. The EO does not include licensing or liability provisions, which could limit openness in AI development.
  3. The EO emphasizes defense against malicious AI uses, registration and reporting requirements, and transparency audits to ensure security and accountability.

The binary sphere packing problem

A Piece of the Pi: mathematics explained • 54 implied HN points • 12 Jul 25
🔬 Science Research
  1. The best way to pack spheres to use space efficiently is known thanks to a theory called the Kepler conjecture. It shows that no arrangement can be denser than stacking spheres in a certain structured way.
  2. When packing two types of spheres together, it’s possible to fill more space than just using one size. An ideal ratio of the sizes can help maximize how much space is used.
  3. Researchers are still working on the binary packing problem to determine how densely two sizes of spheres can fill space. They have found hints that a specific size ratio might help achieve the best packing.

German Newspaper Forces Correction at Science Magazine, While Americans Ignore Study That Undermines Market Theory for Pandemic Origin

The DisInformation Chronicle • 375 implied HN points • 15 Feb 24
🔬 Science Research
  1. A German newspaper forced Science Magazine to correct a study about the pandemic origin, while American science writers ignored new research questioning the study's validity.
  2. The Science Magazine study, claiming the pandemic began in a wet market, was criticized for its statistical methodology by experts from Germany and Hong Kong, raising doubts about its conclusions.
  3. Independent experts confirmed the criticism of the study, highlighting flaws in the statistical analysis and describing Science Magazine's handling of the methodology as careless and unprofessional.

Deciding the Next Research Items

DeFi Education • 779 implied HN points • 12 Aug 21
🔮 Crypto Research
  1. The next research topic will be about Sushi, and it's happening soon. It's a way to keep readers informed and engaged.
  2. Free subscribers can now suggest research projects, making everyone's voice heard. This helps the community feel involved in the content.
  3. Requests from paid subscribers get priority, but everyone's input is valued. It creates a balance between supporting paid content and considering free members' interests.

The Pareto frontier approach to science

Splitting Infinity • 59 implied HN points • 11 Jan 24
🔬 Science Research
  1. Creating new possibilities in science can be more valuable than just focusing on practical or purely exploratory research.
  2. The Pareto Frontier approach in science involves pushing frontiers by inventing solutions that lie at the cutting edge of various parameters.
  3. By extending the frontiers of knowledge in a field, we not only enable practical applications but also broaden the horizons of future innovators.

A minimal viable product for alignment

Musings on the Alignment Problem • 399 implied HN points • 29 Mar 22
🕹 Technology Research
  1. Progress in AI can expand the range of problems humanity can solve, addressing the limitation of human capabilities.
  2. Automating alignment research using AI systems can accelerate progress by overcoming talent bottlenecks and enabling faster evaluation and generation of solutions.
  3. An alignment MVP approach is less ambitious than solving all alignment problems but can still lead to solutions by leveraging automation and AI capabilities.

The RCT Agenda

Bet On It • 352 implied HN points • 11 Mar 24
🚌 Education Research
  1. The book highlights how the randomistas in economics might have a political agenda of serving government rather than challenging it, focusing on randomized controlled trials (RCTs).
  2. Many economists focus on RCTs to measure causal effects of variables, emphasizing laboratory, field, and natural experiments in research, with randomization as the core of experimentation.
  3. Randomistas often avoid engaging in the debate of free markets vs. government, favoring government-centered policies in their research strategies, leading to limited exploration of free-market economics.

Roles of Supervised Machine Learning in Science

Mindful Modeler • 179 implied HN points • 31 Jan 23
🔬 Science Research
  1. Machine learning models play multiple roles in science: as study objects, scientific tools, and scientific models.
  2. Using machine learning models as study objects is common in science, focusing on predictive model performance comparisons.
  3. Machine learning models can be utilized as scientific tools and as scientific models, where they play a central role in understanding phenomena.

Understanding Variation in PTEN

Holodoxa • 99 implied HN points • 07 Sep 23
🔬 Science Research
  1. Understanding genomic data variation and its effect is a significant challenge in genetic research.
  2. Deep Mutational Scanning (DMS) and Multiplex Assays of Variant Effects (MAVEs) are crucial methods to study how mutations impact protein function.
  3. MAVE data on PTEN has provided insights into its function, stability, and clinical implications, aiding in the understanding of PTEN variation.

Enterprise Prompt Engineering Practices

Cobus Greyling on LLMs, NLU, NLP, chatbots & voicebots • 19 implied HN points • 17 May 24
🕹 Technology Research
  1. Users spend a good amount of time, around 43 minutes, editing prompts to get better results from language models. They often make small, careful changes instead of big rewrites.
  2. The main focus of edits is usually on the context of the prompts, such as improving examples and grounding information. This shows that context is crucial for getting good outputs.
  3. Many users try multiple changes at once and sometimes roll back their edits. This indicates that they might struggle to remember what worked well in the past or which changes had positive effects.

Microsoft's New Framework for Multi-Agent Systems

TheSequence • 175 implied HN points • 10 Nov 24
🕹 Technology Research
  1. Magentic-One is a new tool from Microsoft that helps manage multiple AI agents to tackle complex tasks. It acts like a conductor guiding different musicians, making it easier to complete different jobs together.
  2. This system allows for flexibility by using different AI models for different tasks, which means it can be customized based on what you need. It's designed to improve efficiency in our daily tasks, like ordering food or doing research.
  3. While Magentic-One is powerful, it's still being improved to reduce errors and ensure it acts safely. The goal is to make sure these AI agents help us reliably without causing problems.

Erika Update #1: Getting projects started is hard

Erika’s Newsletter • 98 implied HN points • 20 Feb 23
🔬 Science Research
  1. Starting projects can be challenging, but perseverance and a supportive environment are key to making progress.
  2. Mistakes are common in research, but being able to identify and correct them is crucial for success.
  3. Regular updates and reflections on the progress of a project can provide valuable insights and contribute to overall success.

Links for 2023-06-01

Axis of Ordinary • 98 implied HN points • 01 Jun 23
🕹 Technology Research
  1. Model training can be improved by rewarding each correct step of reasoning in mathematical problem solving.
  2. New fMRI-to-image approach called MindEye retrieves and reconstructs images from brain activity.
  3. Probabilistic AI can assess its own performance effectively.

The Last Twitter Show

WORLD GONE WRONG • 98 implied HN points • 11 Mar 23
🇺🇸 U.S. Politics Research
  1. White supremacists have gained millions of followers on social media and have significant influence.
  2. The future of studying social networks like Twitter may become more challenging and costly due to potential changes on the platform.
  3. Despite the negative aspects, Twitter has served as a valuable platform for work, news, and connections.

Google Maps Travel Time Prediction Algorithm & AI Research at Google Deepmind

What's AI Newsletter by Louis-François Bouchard • 98 implied HN points • 09 Jul 23
🕹 Technology Research
  1. The podcast episode discusses Google Maps travel time prediction algorithm and AI research at Google Deepmind.
  2. Petar Veličković shares his journey from academia to developing the algorithm used in Google Maps.
  3. The interview sheds light on opportunities and challenges in the rapidly evolving field of machine learning.

Fallacies in Computational Neuroscience

NeuroLogos • 98 implied HN points • 25 Apr 23
🔬 Science Research
  1. Garbage in, garbage out - common issue in computational models
  2. Unity Gain Simulation - building intricate models of basic concepts without gaining insights
  3. The Prayer Wheel - emphasizing model complexity and need for powerful computers as a form of validation