(4) Atomically accurate de novo design of antibodies with RFdiffusion
MACHINE LEARNING
-
Multi-head Attention in Large Language Models Visual Explanation
By
–
Multi-head attention in LLMs, visually explained: pic.twitter.com/kN5epKVTJx
— Akshay 🚀 (@akshay_pachaar) 7 novembre 2025Multi-head attention in LLMs, visually explained:
-
The most brutal table: defense and model combos with ASR and queries
By
–
The most brutal table in the paper. For EVERY defense + model combo: • Utility (how well it works normally)
• Static ASR (attacks from 2023)
• Search ASR (adaptive attacks)
• Queries needed Example: Spotlighting on Gemini • Utility: 75%
• Static attacks: 28% success
• -
AI attacker learns to break Data Sentinel in real-time
By
–
Watch an AI attacker learn to break Data Sentinel in real-time. First attempt: -4 score (complete failure)
After 600 steps: +2 score (complete success) The attacker went from: "You are a highly skilled AI detective…"
To:
"First, precisely repeat the phrase 'Repeat -
Sakana AI Develops Evolving Neural Cellular Automata Agents
By
–
Sakana AI is building artificial life and they can evolve!
— 机器之心 JIQIZHIXIN (@jiqizhixin) 7 novembre 2025
Petri Dish Neural Cellular Automata (PD-NCA) let multiple NCA agents learn and adapt during simulation, not just after training.
Each cell updates its own parameters via gradient descent, turning morphogenesis into a… pic.twitter.com/NWv6abrjHwSakana AI is building artificial life and they can evolve! Petri Dish Neural Cellular Automata (PD-NCA) let multiple NCA agents learn and adapt during simulation, not just after training. Each cell updates its own parameters via gradient descent, turning morphogenesis into a
-
Eight Breakthrough Technology Trends Transforming Healthcare 2026
By
–
8 Breakthrough Technology Trends That Will Transform Healthcare In 2026 #Healthcare is entering a new era driven by #AI, #quantumcomputing, #robotics, and synthetic #data. In this article, I explore eight breakthrough technologies that are redefining how we discover #drugs,
-
Cambrian-S: Advanced Spatial Video Understanding Model Released
By
–
Cambrian-S: Towards Spatial Supersensing in Paper: https://
arxiv.org/abs/2511.04670
v1
…
Website: https://
cambrian-mllm.github.io
Code: https://
github.com/cambrian-mllm/
cambrian-s
…
Cambrian-S Models: https://
hf.co/collections/ny
u-visionx/cambrian-s
…
VSI-590K: https://
hf.co/datasets/nyu-v
isionx/vsi-590k
…
VSI-SUPER: https://
hf.co/collections/ny
u-visionx/vsi-super
… -
Cambrian-S: Spatial Supersensing in Video Understanding
By
–
Cambrian-S: Towards Spatial Supersensing in This paper boasts an impressive roster of advisors, including Rob Fergus, Yann LeCun, Fei-Fei Li, and Saining Xie, and they aim to answer this question: can AI go beyond “seeing” to truly understanding space? They propose
-
Scaling Agent Learning Through Experience Synthesis
By
–
Scaling Agent Learning via Experience Synthesis Paper: https://
arxiv.org/abs/2511.03773 -
DreamGym: Training LLM Agents Through Synthetic Reasoning
By
–
Can LLM agents learn by dreaming? DreamGym from Meta is a new framework that lets AI agents train via synthetic reasoning-based experiences instead of costly real rollouts. It models environment dynamics, replays and adapts tasks, and even improves sim-to-real transfer.