High-speed AFM imaging reveals how brain enzyme forms a dodecameric ring structure

Kanazawa University, have captured real-time images showing how a key brain enzyme organizes itself to help memory formation.
MIT Technology Review

Mechanistic interpretability

But last year we got the best sense yet of how LLMs function, as researchers at top AI companies began developing new ways to ...
MIT Technology Review

Meet the new biologists treating LLMs like aliens

By studying large language models as if they were living things instead of computer programs, scientists are discovering some ...

A Visual Model Of Self-Attention: Transformers Work Differently Now

Early-2026 explainer reframes transformer attention: tokenized text becomes Q/K/V self-attention maps, not linear prediction.
GitHub

MAESTRO: Masked Autoencoders for Multimodal, Multitemporal, and Multispectral Earth Observation Data

MAESTRO_FLAIR-HUB_base — pre-trained on FLAIR-HUB MAESTRO_S2-NAIP-urban_base — pre-trained on S2-NAIP-urban Land cover segmentation in France, with 12 semantic classes. Note that the FLAIR#2 version ...
IEEE

Anomaly Detection in IoT Networks: A Deep Learning Approach Using Autoencoders

Abstract: The current fast proliferation of the Internet of Things (IoT) networks has made anomaly detection and security more difficult. Traditional methods are not able to detect hostile activities ...
IEEE

Integrating Contractive Loss in Proximal Policy Optimization for Robust Decision-Making

Abstract: Deep Reinforcement Learning (DRL) has found successful applications across various domains, including robotics, healthcare, finance, and autonomous systems. However, in real-world ...
GitHub

AVF-MAE++ : Scaling Affective Video Facial Masked Autoencoders via Efficient Audio-Visual Self-Supervised Learning

Abstract: Affective Video Facial Analysis (AVFA) is important for advancing emotion-aware AI, yet the persistent data scarcity in AVFA presents challenges. Recently, the self-supervised learning (SSL) ...