Blog

I often test new LLM models with the following question — Given the following conditions, how many ways can Professor Y assign six different books to four different students? The most expensive book must be assigned to student X. Each student must receive at least one book. The first model which can solve this problem is OpenAI’s o1 model. The second one is Google’s Gemini 2.0 Flash Thinking model. Then more of such thinking models emerged, such as DeepSeek’s R1, Anthropic’s Claude 3.7 (with extended thinking mode), all of which can solve this problem flawlessly. ...

A robot valet wondering in doomed world Adrian Tchaikovsky, the science fiction writer, is a master at crafting bleak, hellish future worlds. But in Service Model, a 2024 science-fiction satire telling the story of a robot valet wandering in a doomed world, he has truly outdone himself, conjuring an absurd realm where human societies have crumbled, and humanity teeters on the brink of extinction. ...

A friend ask me if I find any meaning in my work. Here’s my reply. Please forgive me for answering your question in English. I find it easier to express myself clearly in English, especially on more complex topics. You ask if I find it “meaningful” to write papers that few people might read. The short answer is “Yes.” The long answer, as usual, is more complicated. Will my research cure cancer, bring world peace, or entertain millions? No. Will it contribute to a technological breakthrough one day? It’s unlikely, though possible. ...

A scene in Here (2023) Here (2023), directed by Bas Devos, is a poetic meditation on the everyday. Eschewing traditional dramatic structures, the film flows with the quiet rhythm of life itself. The story follows Stefan, a Romanian construction worker in Brussels, who, on the verge of returning home, meets Shuxiu, a Belgian-Chinese bryologist studying moss—those tiny, verdant worlds often overlooked beneath our feet. Their chance encounter sets off a contemplative exploration of the city’s hidden beauty, reminding us that wonder often resides in the unlikeliest of places. ...

My favourite philosopher, Peter Singer, has, like many in today’s world, ventured into the realm of podcasting with Live Well Lived, alongside his collaborator, Kasia de Lazari Radek. Together, they have engaged in conversations with experts from diverse fields. For instance, they spoke with psychiatrist Murali Doraiswamy about tackling global mental health challenges. I found the following exchange both succinct and remarkably pertinent to the podcast’s theme — Kasia de Lazari Radek: Okay, so now I should ask you, do you believe that we can train our brain to be happier on its own without any drugs, any psychedelics and any psychiatric drugs? ...

A poster for the film Perfect Days On some dating websites, you might find yourself confronting that deceptively profound question: “What is a perfect day for you?” Like most seemingly simple queries about the human condition, its answers are as numerous and varied as the stars. For Ayelet Waldman, author of A Really Good Day, a perfect day arrives through carefully measured doses of LSD — a chemical key to unlock the door to contentment. In contrast, the 2023 film Perfect Days, directed by Wim Wenders and starring Kōji Yakusho, perfection whispers in a different language: it speaks through the sacred ordinary of each passing day. ...

I recently learned, from the ever-illuminating Simon Willison, that Google has released a new model, gemini-2-0-flash-thinking-exp. So, naturally, I posed it a question, a little puzzle if you will. Given the following conditions, how many ways can Professor Y assign six different books to four different students? The most expensive book must be assigned to student X. Each student must receive at least one book. It took the model some time, a moment of digital contemplation, to complete the task. But it did, correctly yielding the answer 390. So now, alongside gpt-o1, we have another model capable of solving this combinatorial problem, a problem that, it must be said, eludes most of my students. It seems that these “inference scaling” based models will be able to solve more such problems, each one a small victory for the silicon mind. Another reason, perhaps, why I no longer set homework and only give quizzes, a small act of surrender in the face of the inevitable. ...

I just finished The Mercy of Gods by James S.A. Corey, best known for the Expanse series. It’s a space opera about a team of top biologists who find themselves taken captive by brutal alien invaders. Forced into servitude, they must decide how to navigate life under these new, hostile conditions. The setup is rich with tension and moral dilemmas. The first part of the book reveals certain dynamics in the academic world — competition for discoveries, power plays, and the undercurrent of personal ambitions. Even in pursuit of something noble, like advancing knowledge, human frailties often take centre stage. ...

🙏 Acknowledgement I recently came across a thought-provoking post by Benjamin Hilton of 80,000 Hours on the Effective Altruism Forum. The post provides a detailed overview of the immense harm caused by factory farming. Below is an excerpt highlighting some of the most shocking facts about the industry. For detailed references, please refer to the original article. This excerpt is also posed on Plant Futures DKU’s website. A pig confined to a pen. Photo by Matthias Zomer on Pexels. ...

🙏 Acknowledgement This essay by George Monbiot, a columnist for The Guardian, is part of his book This Can’t Be Happening, a collection of his writings on environmental issues. We are grateful to Monbiot for allowing us to translate it to Chinese and republish his essay here. We Can Stop the First Great Extermination George Monbiot October 2019 The Launch of Animal Rebellion, Smithfield Market, London I have spent my life learning and unlearning. We cannot navigate the world in which we live without unlearning some of its deepest myths. ...