Rapid7 dropped a write-up on the Notepad++ update-chain abuse and - finally - it comes with real IOCs - update.exe downloaded from 95.179.213[.]0 after notepad++.exe -> GUP.exe - file hashes for update.exe / log.dll / BluetoothService.exe / conf.c / libtcc.dll - network IOCs

Our new paper discusses the process of applying an AI math research agent built on top of Deep Think to unsolved Erdős problems. We identified 13 correct responses, 8 from existing literature and 5 representing novel autonomous solutions. arxiv.org/abs/2601.22401

Excited to share an extension to spectral filtering for the complex domain, including quantum dynamical systems, w. my former postdoc Annie M.: arxiv.org/abs/2601.22400

[LG] Statistical Estimation of Adversarial Risk in Large Language Models under Best-of-N Sampling M Feng, X Liu, W Yang, C Xu... [Microsoft Research] (2026) arxiv.org/abs/2601.22636

🔥 Excited to share RLAnything (arxiv.org/abs/2602.02488): A fully dynamic RL framework that jointly forges Environment, Policy, and Reward Model through closed-loop optimization! 🧵 Let's talk about the problems first: ❌ Problem 1: Sparse Rewards Current RLVR relies on binary

Beyond RAG for Agent Memory: Retrieval by Decoupling and Aggregation Introduces a hierarchical memory framework that disentangles agent memory into semantic components and retrieves top-down, reducing redundancy while improving answer quality. 📝 arxiv.org/abs/2602.02007

Today we’re releasing the International AI Safety Report 2026: the most comprehensive evidence-based assessment of AI capabilities, emerging risks, and safety measures to date. 🧵 (1/17)

Wrong question: The assumption that conversational language use can only be a sign of intelligence--i.e. that speech is predicated on an intelligent speaking subject--is the operating system of logocentric metaphysics. And it is deconstructed here: link.springer.com/article/10.100…

🧠 Why do LLMs fail? Sometimes the failure is what we expect - but often, it isn’t. Curious about overlooked but common LLM failures? Want to debug your recent fine-tuned model? Checkout our paper: arxiv.org/abs/2601.15812 #LLM #AI #NLP #MachineLearning #error_analysis

Moltbook is just the beginning. Social Media Is a Warzone. Where’s Your Armor? OSINT stands for Open Source Intelligence: the collection and analysis of info to produce actionable intelligence. Building your own OSINT defenses isn’t optional anymore. garryslist.org/posts/social-m…

🚨 Hopefully, AI companies will give up on their highly anthropomorphic, emotionally manipulative AI chatbots and focus on science-oriented, specialized AI applications like this. A group of MIT researchers trained an AI model on 23,000+ material synthesis recipes described in

Today we’re launching Phylo, a research lab studying agentic biology, backed by a $13.5M seed round co-led by a16z and Menlo Ventures / Anthology Fund Anthropic. We’re also introducing a research preview of Biomni Lab, the first Integrated Biology Environment (IBE), where

Parallel Stochastic Gradient-Based Planning for World Models Michael Psenka, Michael Rabbat, Aditi Krishnapriyan, Yann LeCun, Amir Bar arxiv.org/abs/2602.00475 [𝚌𝚜.𝙻𝙶 𝚌𝚜.𝚁𝙾]

This study exemplifies #AI's power to help researchers solve highly complex scientific problems. Physical Review B bnl.gov/newsroom/news.…

AI agent social network Moltbook grew from 30,000 users on January 30 to 1.5M+ on February 2; researchers say many viral posts were likely scripted by humans (Hayden Field / The Verge) theverge.com/ai-artificial-… techmeme.com/260203/p25#a26… 📥 Send tips! techmeme.com/contact

x.com/i/article/2018…

Dispelling the Curse of Singularities in Neural Network Optimizations Hengjie Cao, Mengyi Chen, Yifeng Yang, Fang Dong, Ruijun Huang, Anrui Chen, Jixian Zhou, Mingzhi Dong, Yujiang Wang, Dongsheng Li, Wenyi Fang, Yuanyi Lin, … arxiv.org/abs/2602.01308 [𝚌𝚜.𝙻𝙶 𝚌𝚜.𝙰𝙸]

[CL] Training LLMs for Divide-and-Conquer Reasoning Elevates Test-Time Scalability X Liang, Z Li, Z Lin, E H Jiang... [University of California, Los Angeles & Microsoft] (2026) arxiv.org/abs/2602.02477

Measuring Cyber Risk in the Italian Corporate Sector – Random Bits of Knowledge 4m4.it/posts/measurin…

Measuring Cyber Risk in the Italian Corporate Sector 4m4.it/posts/measurin…