September
Papers
- Vision Transformers Need Registers
- π0: A Vision-Language-Action Flow Model for General Robot Control
Links
- https://arxiv.org/abs/2608.28444 Sliding-window beats linear attention
- https://arxiv.org/abs/2502.17416 Reasoning with Latent Thoughts: On the Power of Looped Transformers
- https://arxiv.org/abs/2511.05963 Next-Latent Prediction Transformers Learn Compact World Models
- https://mostik.ai/read-more Bridging models’ internal states
- https://www.anthropic.com/research/natural-language-autoencoders Natural Language Autoencoders: Turning Claude’s thoughts into text
- https://blog.google/innovation-and-ai/technology/research/male-fruit-fly-brain-map/ 5 amazing visuals show how the male fruit fly’s brain map is advancing neuroscience
- https://arxiv.org/abs/2609.02737 Language Models Can Control Their Own Attention
- https://sebastianraschka.com/llm-architecture-gallery/ LLM Architecture Gallery
- https://xenaproject.wordpress.com/2026/09/04/flt-anthropic-has-beaten-me-to-it/ FLT: Anthropic has beaten me to it
- https://collusion.wiki/ Discovery of a new OpenAI agent message board
- https://austinhenley.com/blog/python1024.html Making a Python interpreter in 1024 bytes
- https://www.lesswrong.com/posts/PLisnSFir8y5AHkmP/how-concerned-should-we-be-about-astra-s-recurrent How concerned should we be about Astra's recurrent architecture?
- https://magazine.sebastianraschka.com/p/gpt-6-astra-looped-transformers-and GPT-6 Astra, Looped Transformers, and Hidden Reasoning
- https://hugovergnes.github.io/little-lm-3-8b/ Training a 3.8B LLM to 0.384 CORE for $998
- https://erich-friedman.github.io/packing/ Packing Equal Copies
- https://sites.research.google/gr/neural-mapping/ Neural Mapping - Google Research
- https://mathandai.org/ A Severe Misalignment of AI in Mathematics
- https://journal.stuffwithstuff.com/2015/02/01/what-color-is-your-function/ What Color is Your Function?
- https://www.pi.website/blog/pi0 π0: Our First Generalist Policy
- https://arxiv.org/abs/2303.15343 Sigmoid Loss for Language Image Pre-Training
- https://x.com/mayukh_panja/status/2098407452630770095 No, OpenAI Did Not “Solve Navier–Stokes.”
- https://arxiv.org/abs/2609.11873 The Last AI Built by Humans: Toward Genuine Recursive Self-Improvement
- https://www.vals.ai/blogs/fable-solves-cyphral-distich Claude Fable 5.1 Solves the Cyphral Distich
- https://github.com/adonis-singh/TMNF-C TMNF-C
- https://pub.sakana.ai/pc-alm/ Augmented Lagrangian Predictive Coding - Training 1000-layer networks without backpropagation
- https://github.com/kevinzakka/mjbatch mjbatch - a Python library for running thousands of MuJoCo simulations in parallel on CPU.
- https://pmc.ncbi.nlm.nih.gov/articles/PMC12629191/ 3D necroprinting: Leveraging biotic material as the nozzle for 3D printing
- https://improbable.com/ig/winners/?amp=1 Past Nobel Prize Winners
- https://prismml.com/news/bonsai-2-27b Introducing Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint
- https://gowers.wordpress.com/2026/09/17/why-i-didnt-sign-the-fields-medallists-letter/ Why I didn’t sign the Fields medallists’ letter
- https://huggingface.co/harshatheg/Qwen-2.5-1B-RLCD Parallel Constrained Decoding for Apple Silicon
- https://elifesciences.org/articles/10778 Nanoconnectomic upper bound on the variability of synaptic plasticity
- https://terrytao.wordpress.com/2026/09/18/if-math-is-more-than-proof-we-need-to-better-celebrate-the-rest-of-it/ If math is more than proof, we need to better celebrate the rest of it
- https://john.hartnup.uk/2026/06/07/ai-event-posters.html AI-generated posters don’t have to be horrible
- https://cast.ai/blog/demystifying-quantizations-llms/ Demystifying Quantizations: Guide to Quantization Methods for LLMs
- https://k4i.top/posts/llm-quantization/ A Survey of LLM Quantization: From Linear Quantization to Codebooks
- https://mimo.xiaomi.com/rl/ Mimo-v2.6 live training metrics
- https://mimo.xiaomi.com/mimo-v2-6 Introducing MiMo-V2.6 series
- https://kokosa.dev/blog/2026/dotfly/ dotFly - a fruit fly's brain, simulated in C#
- https://blog.pioneer-labs.org/p/announcing-the-first-microbe-for Announcing the First Microbe for Mars
- https://whitecircle.com/halo Halo: Frontier-Lab Training for Everyone
- https://x.com/theSamPadilla/status/2102533979291443495 Robotics Data is a Broken Business
- https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-8-text-to-speech/ Gemini 3.8 text-to-speech says hello
- https://www.quantamagazine.org/updates/transformation/#167384 Software developer finds first 3D “einstein” tile
- https://contrastive-lm.notion.site/ Contrastive Language Models
- https://di-zhang-llm.github.io/blog/what-is-rlcd-the-secret-behind-jev/ What Is RLCD? The Secret Behind Jev
- https://worksinprogress.co/issue/weapons-of-mass-decentralization/ Weapons of mass decentralization
- https://users.cs.northwestern.edu/~robby/uc-courses/22001-2008-winter/scrabble.html Scrabble Rules
- https://mathigon.org/origami Mathematical Origami