#cloud

Thinking about the recent aws outage, this classical piece of work is still relevant after so many years. The paradox of automation, or cloud infra, is so real.

Ironies+of+Automation_Bainbridge_1983.pdf

615.4 KB

#fun

https://www.reddit.com/r/learnmachinelearning/s/HKIdbsJYXc

Reddit

From the learnmachinelearning community on Reddit: [D] Can someone please teach me how transformers work? I heard they are used…

Explore this post and more from the learnmachinelearning community

#ai

Souly, Alexandra, Javier Rando, Ed Chapman, Xander Davies, Burak Hasircioglu, Ezzeldin Shereen, Carlos Mougan, et al. 2025. “Poisoning Attacks on LLMs Require a Near-Constant Number of Poison Samples.” arXiv [Cs.LG]. arXiv. http://arxiv.org/abs/2510.07192.

arXiv.org

Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samples

Poisoning attacks can compromise the safety of large language models (LLMs) by injecting malicious documents into their training data. Existing work has studied pretraining poisoning assuming...

#dl

Park, Chanwook, Sourav Saha, Jiachen Guo, Hantao Zhang, Xiaoyu Xie, Miguel A. Bessa, Dong Qian, et al. 2025. “Unifying Machine Learning and Interpolation Theory via Interpolating Neural Networks.” Nature Communications 16 (1): 1–12.
https://www.nature.com/articles/s41467-025-63790-8

#ai

OpenAI open weight models. The benchmarks are amazing.

https://huggingface.co/collections/openai/gpt-oss-68911959590a1634ba11c7a4

#misc

How we learned to stop worrying and love ChatGPT: 1,800 researchers on AI tools for science - Paperpile
https://paperpile.com/blog/ai-tools-for-science/

Paperpile

How we learned to stop worrying and love ChatGPT: 1,800 researchers on AI tools for science - Paperpile

Read about Paperpile’s data-driven approach to learn about the AI tools and use cases that expert researchers value most. Paperpile incorporated this survey feedback into their forward-looking AI roadmap, including a key set of principles to guide development.

#misc

One year later, raycast for Windows is finally in beta.

#misc

So stupid if this is true.


Jordi Hays on X: "Here is most of what I’ve gathered on the Windsurf / Google Deal The founders and dozens of engineers are going to Google. This group, along with the preferred shareholders will be sharing the $2.4B headline number. The exact split is unknown but investors are making some money" / X
https://x.com/jordihays/status/1944200891944644997

X (formerly Twitter)

Jordi Hays (@jordihays) on X

Here is most of what I’ve gathered on the Windsurf / Google Deal

The founders and dozens of engineers are going to Google. This group, along with the preferred shareholders will be sharing the $2.4B headline number. The exact split is unknown but investors…

#ai

PedanticAI already partially supports A2A .

Agent2Agent (A2A) Protocol
https://a2aproject.github.io/A2A/dev/

a2aproject.github.io

Agent2Agent (A2A) Protocol

An open protocol enabling communication and interoperability between opaque agentic applications.

#dl

A few cool ideas in this model.

Introducing Gemma 3n: The developer guide - Google Developers Blog
https://developers.googleblog.com/en/introducing-gemma-3n-developer-guide/

Googleblog

Google for Developers Blog - News about Web, Mobile, AI and Cloud

Learn how to build with Gemma 3n, a mobile-first architecture, MatFormer technology, Per-Layer Embeddings, and new audio and vision encoders.

#ml

Machine Learning Visualized — Machine Learning Visualized
https://ml-visualized.com/?utm_source=substack&utm_medium=email

#dl

So tensorflow and jax are deprecated in the transformers package.

https://github.com/huggingface/transformers/pull/38758

GitHub

Deprecate TF + JAX by Rocketknight1 · Pull Request #38758 · huggingface/transformers

The time has finally come 🔫 🥃

#ai

Well, I figure many people actually "think" like this.

The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity
https://ml-site.cdn-apple.com/papers/the-illusion-of-thinking.pdf

#fun

The result looks good. It probably makes better choices than me.

Lopez-Lira, Alejandro, and Yuehua Tang. 2023. “Can ChatGPT Forecast Stock Price Movements? Return Predictability and Large Language Models.” arXiv [q-Fin.ST]. arXiv. http://arxiv.org/abs/2304.07619.

arXiv.org

Can ChatGPT Forecast Stock Price Movements? Return Predictability...

We document the capability of large language models (LLMs) like ChatGPT to predict stock market reactions from news headlines without direct financial training. Using post-knowledge-cutoff...

Mary Meeker's first Trends report since 2019. 340 slides on the state of AI.
https://www.bondcap.com/reports/tai

#ai

Extending Minds with Generative AI | Nature Communications
https://www.nature.com/articles/s41467-025-59906-9

Nature

Extending Minds with Generative AI

Nature Communications - As human-AI collaborations become the norm, we should remind ourselves that it is our basic nature to build hybrid thinking systems – ones that fluidly incorporate...

#forecasting

Hyndman's new book

https://otexts.com/fpppy/

Otexts

Forecasting: Principles and Practice, the Pythonic Way