January
Papers
- It Takes Two Neurons To Ride a Bicycle
- Can LLMs reason over extended multilingual contexts? Towards long-context evaluation beyond retrieval and haystacks
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Links
- https://simonwillison.net/2025/Dec/31/the-year-in-llms/ 2025: The year in LLMs
- https://sakana.ai/ahc058/ Sakana AI Agent Wins AtCoder Heuristic Contest (First AI to Place 1st)
- https://blog.booleanbiotech.com/oral-microbiome-biogaia Oral microbiome sequencing after taking probiotics
- https://www.k-a.in/RLM.html Recursive Language Models
- https://pragmaticproteindesign.bio.ed.ac.uk/pda/ The Protein Design Archive
- https://www.kaggle.com/competitions/stanford-rna-3d-folding-2 Stanford RNA 3D Folding Part 2
- https://www.owlposting.com/p/a-2026-look-at-three-bio-ml-opinions A 2026 look at three bio-ML opinions I had in 2024
- https://boltz.bio/boltzlab Announcing Boltz Lab and our first agents
- https://sakana.ai/drq/ Digital Red Queen: Adversarial Program Evolution in Core War with LLMs
- https://c4ss.org/content/50557 Widening the Bridges: Beyond Consent and Autonomy
- https://www.cs.helsinki.fi/u/ahyvarin/painintl/ Painful intelligence: What AI can tell us about human suffering
- https://www.toposbio.ai/tech-report Introducing Topos-1
- https://www.erdosproblems.com/ Erdos Problems
- https://aresluna.org/fav-tech-museums/ Fav tech museums
- https://zimmer.csufresno.edu/~larryc/proofs/proofs.html How To Write Proofs
- https://centuryofbio.com/ The Century of Biology
- https://pub.sakana.ai/DroPE/ Extending the Context of Pretrained LLMs by Dropping their Positional Embeddings
- https://github.com/deepseek-ai/Engram Deepseek Engram
- https://bricknerd.com/home/going-for-gold-the-story-of-the-golden-lego-rcx-and-nxt-9-9-21 Going for Gold: The Story of the Golden RCX and NXT
- https://certificate.transparency.dev/logs/ The list of current, usable logs
- https://seedfold.github.io/ SeedFold - Scaling Biomolecular Structure Prediction
- https://openinframap.org/#2.39/35.84/110.92/A,B,L,P,S Open Infrastructure Map
- https://blog.google/innovation-and-ai/technology/developers-tools/translategemma/ TranslateGemma: A new suite of open translation models
- https://ericjmichaud.com/quanta/ On neural scaling and the quanta hypothesis
- https://nanonets.com/cookbooks/structured-llm-outputs Structured LLM outputs
- https://geohot.github.io/blog/jekyll/update/2026/01/17/three-minutes.html you have three minutes to escape the perpetual underclass
- https://geohot.github.io//blog/jekyll/update/2026/01/18/how-do-i-stop.html how do I stop participating?
- https://en.wikipedia.org/wiki/TRIZ TRIZ
- https://www.inf.ufpr.br/renato/profession.html Profession - Isaac Asimov
- https://github.com/teorth/erdosproblems/wiki/AI-contributions-to-Erd%C5%91s-problems AI contributions to Erdős problems
- https://blog.yakkomajuri.com/blog/raising-money-fucked-me-up Raising money fucked me up
- https://alexharri.com/blog/ascii-rendering ASCII characters are not pixels: a deep dive into ASCII rendering
- https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1012754 The calcitron: A simple neuron model that implements many learning rules via the calcium control hypothesis
- https://bitchat.free/ bitchat
- https://arxiv.org/abs/2601.10825 Reasoning Models Generate Societies of Thought
- https://www.cell.com/current-biology/fulltext/S0960-9822(25)01597-0 Flexible use of a multi-purpose tool by a cow
- https://www.anthropic.com/research/assistant-axis The assistant axis: situating and stabilizing the character of large language models
- https://www.hit.bme.hu/~papay/edu/DSOdisp/gradient.htm DSO/DPO: Gradient - displaying the Data
- https://www.linuxfromscratch.org/lfs/view/stable/ Linux From Scratch
- https://waitbutwhy.com/2014/09/but-what-about-greenland.html But What About Greenland?
- https://dgerrells.com/blog/how-fast-is-javascript-simulating-20-000-000-particles How fast is javascript? Simulating 20,000,000 particles
- https://nesbitt.io/2025/12/26/how-uv-got-so-fast.html How uv got so fast
- https://eieio.games/blog/ssh-sends-100-packets-per-keystroke/ Why does SSH send 100 packets per keystroke?
- https://catonmat.net/tcp-cork What is TCP_CORK?
- https://design-a-protein.com/ Design_a_Protein
- https://rootkid.me/works/exhibit-a Exhibit A (2025)
- https://www.kerneloptimization.fun/ Kernel Optimization Challenge - Can You Beat Claude?
- https://github.com/martinpacesa/BindCraft BindCraft
- https://waitbutwhy.com/2014/05/life-weeks.html Your Life in Weeks
- https://geohot.github.io/blog/jekyll/update/2022/02/09/the-problem-of-the-state.html The Problem of the State
- https://boxkitemachine.net/posts/zero-to-one-peter-thiel-definite-vs-indefinite-thinking/ Definite vs. indefinite thinking: Notes from Zero to One by Peter Thiel
- https://trirpi.github.io/posts/anthropic-performance-takehome/ Deep Dive: Anthropic's Performance Take-Home (The One Claude Beat Humans At)
- https://medium.com/@indosambhav/my-journey-through-the-anthropic-performance-optimization-challenge-7a5dc46dd6e0 My Journey through the Anthropic performance optimization challenge
- https://www.nature.com/articles/s41467-026-68698-5 The network architecture of general intelligence in the human connectome
- https://github.com/google-deepmind/alphagenome_research AlphaGenome Research
- https://www.asimov.press/p/brains Building Brains on a Computer
- https://github.com/karpathy/nanochat/discussions/481 Beating GPT-2 for <<$100: the nanochat journey