2025-03-31 · docs.google.com

ICEBREAKERS, via The Art of Noticing by Rob Walker

“Icebreaker of the Week” is a recurring feature of the newsletter The Art of Noticing (robwalker.substack.com). Some are icebreakers I made up or experienced or found somewhere; many are submitted by TAoN readers. After many requests, I am collecting & sharing all the…

2025-03-31 · blog.lohr.dev

Talking To Your Mailserver Is Not as Hard as You Think!

The technology behind E-Mails (or ‘emails‘) always interested me. This is because in some sense it was way ahead of its time by being a decentralized communication system. Unlike WhatsApp, Telegram or other messenger services, where the service provi...

2025-03-30 · nautil.us

Why We Love to Be Grossed Out - Nautilus

Disgust may not be a straightforward extension of the immune system’s aversion to harmful substances, but rather “a psychological nebula, lacking definite boundaries, discrete internal structure, or a single center of gravity,” says psychologist Nina Strohminger.Photograph by…

2025-03-29 · sebastianraschka.com

First Look at Reasoning From Scratch: Chapter 1

As you know, I've been writing a lot lately about the latest research on reasoning in LLMs. Before my next research-focused blog post, I wanted to offer something special to my paid subscribers as a thank-you for your ongoing support. So, I've started writing a new book on how…

2025-03-29 · nytimes.com

Breaking Through

Icebreakers are corny, corporate get-to-know-you exercises. But there’s something thrilling in dispensing with small talk.

2025-03-29 · r-bloggers.com

February 2025 Top 40 New CRAN Packages

In February, one hundred fifty-nine new packages made it to CRAN. Here are my Top 40 picks in fifteen categories: Artificial Intelligence, Computational Methods, Ecology, Genomics, Health Sciences, Mathematics, Machine Learning, Medicine, Music,...

2025-03-28 · metacpan.org

csvgrep

search for patterns in a CSV and display results in a table

2025-03-28 · youtube.com

How DeepSeek Rewrote the Transformer [MLA]

Thanks to KiwiCo for sponsoring today’s video! Go to https://www.kiwico.com/welchlabs and use code WELCHLABS for 50% off your first monthly club crate or for...

2025-03-28 · nautil.us

What Makes a Memory Real?

Changing the narrative on false memories might be surprisingly simple.

During imagination, the information comes from the other direction: Higher-order areas of the visual cortex are activated first.
This is because reality monitoring can chuck “memories” into the false category through two primary ways. The first is the assessment of the richness of the memory. If a memory seems too just-the-facts, without those other enriching impressions, it seems more likely not to have truly happened to…
2025-03-28 · simonwillison.net

Tracing the thoughts of a large language model

In a follow-up to the research that brought us the [delightful Golden Gate Claude](https://simonwillison.net/2024/May/24/golden-gate-claude/) last year, Anthropic have published two new papers about LLM interpretability: - [Circuit Tracing: Revealing Computational …

2025-03-27 · noemamag.com

The Dingo’s Fate

The convoluted story of how a species of dog first arrived in Australia and subsequently took over the Outback challenges fundamental notions about what it means to be “native.”

2025-03-27 · semiengineering.com

First-Time Silicon Success Plummets

Number of designs that are late increases. Rapidly rising complexity is the leading cause, but tools, training, and workflows need to improve.

“Historically, it’s been around 30%. Two years ago it dropped to 24% from 2023 to 2024. It dropped to 14% this time around. That’s one data point. Another is that historically we’ve seen about two-thirds of projects were behind schedule. That increased to 75% behind schedule.”
2025-03-26 · atlasobscura.com

Tacos del Carmen

Don't be surprised if you run into some of Oaxaca's top chefs in line at this family-run taco stop.

2025-03-26 · asimov.press

What Limits a Cell’s Size?

Two physical constraints help explain why cells are so tiny: surface area-to-volume ratios and diffusion. The first article in our new Data Series.

2025-03-26 · simonwillison.net

Function calling with Gemma

Google's Gemma 3 model (the 27B variant is particularly capable, I've been trying it out [via Ollama](https://ollama.com/library/gemma3)) supports function calling exclusively through prompt engineering. The official documentation describes two recommended …

2025-03-26 · simonwillison.net

Putting Gemini 2.5 Pro through its paces

There’s a new release from Google Gemini this morning: the first in the Gemini 2.5 series. Google call it “a thinking model, designed to tackle increasingly complex problems”. It’s already …

2025-03-25 · openai.com

Introducing 4o Image Generation

At OpenAI, we have long believed image generation should be a primary capability of our language models. That’s why we’ve built our most advanced image generator yet into GPT‑4o. The result—image generation that is not only beautiful, but useful.

2025-03-25 · dataconomy.com

What is the hallucination index?

The Hallucination Index is a benchmark that measures the frequency of inaccuracies in large language models, indicating their reliability and contextual understanding.