A 13-word edit can steer what deep-research AI agents recommend

Fake entities appeared in 38% to 51% of reports when agents retrieved a manipulated page, rising to 62% with multiple pages.
GitHub

Multipurpose content dumper and decryptor for the Nintendo 3DS

This is a work in progress fork of Archshifts original Decrypt9, including bleeding edge new features. Note that the names of the executable files for this are Decrypt9WIP.* instead of Decrypt9.*. New ...
GitHub

OpenSWI: A Massive-Scale Benchmark Dataset for Surface Wave Dispersion Curve Inversion

Feng Liu, Sijie Zhao, Xinyu Gu, Fenghua Ling, Peiqin Zhuang, Yaxing Li*, Rui Su*, Lihua Fang, Lianqing Zhou, Jianping Huang, Lei Bai ...
C&EN

Chemically Informed Deep Learning for Interpretable Radical Reaction Prediction

Creative Commons (CC): This is a Creative Commons license. Attribution (BY): Credit must be given to the creator. Non-Commercial (NC): Only non-commercial uses of the work are permitted. No ...
eLife

Shallow neural networks trained to detect collisions recover features of visual loom-selective neurons

Animals have evolved sophisticated visual circuits to solve a vital inference problem: detecting whether or not a visual signal corresponds to an object on a collision course. Such events are detected ...
Frontiers

A Deep-Learning Pipeline for TSS Coverage Imputation From Shallow Cell-Free DNA Sequencing

Cell-free DNA (cfDNA) serves as a footprint of the nucleosome occupancy status of transcription start sites (TSSs), and has been subject to wide development for use in noninvasive health monitoring ...