Menu

Blog

Page 203

Apr 5, 2024

RNA Molecules in Brain Nerve Cells Display Lifelong Stability

Posted by in categories: biotech/medical, genetics, life extension, neuroscience

Certain RNA molecules in the nerve cells in the brain last a life time without being renewed. Neuroscientists from Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) have now demonstrated that this is the case together with researchers from Germany, Austria and the USA. RNAs are generally short-lived molecules that are constantly reconstructed to adjust to environmental conditions. With their findings that have now been published in the journal Science, the research group hopes to decipher the complex aging process of the brain and gain a better understanding of related degenerative diseases.

Most cells in the human body are regularly renewed, thereby retaining their vitality. However, there are exceptions: the heart, the pancreas and the brain consist of cells that do not renew throughout the whole lifespan, and yet still have to remain in full working order. “Aging neurons are an important risk factor for neurodegenerative illnesses such as Alzheimer’s,” says Prof. Dr. Tomohisa Toda, Professor of Neural Epigenomics at FAU and at the Max Planck Center for Physics and Medicine in Erlangen. “A basic understanding of the aging process and which key components are involved in maintaining cell function is crucial for effective treatment concepts:”

In a joint study conducted together with neuroscientists from Dresden, La Jolla (USA) and Klosterneuburg (Austria), the working group led by Toda has now identified a key component of brain aging: the researchers were able to demonstrate for the first time that certain types of ribonucleic acid (RNA) that protect genetic material exist just as long as the neurons themselves. “This is surprising, as unlike DNA, which as a rule never changes, most RNA molecules are extremely short-lived and are constantly being exchanged,” Toda explains.

Apr 5, 2024

Paper page — CoMat: Aligning Text-to-Image Diffusion Model with Image-to-Text Concept Matching

Posted by in category: robotics/AI

LVLM-Intrepret.

An interpretability tool for large vision-language models.

In the rapidly evolving landscape of artificial intelligence, multi-modal large language models are emerging as a significant area of interest.

Continue reading “Paper page — CoMat: Aligning Text-to-Image Diffusion Model with Image-to-Text Concept Matching” »

Apr 5, 2024

Newly Approved Rapid Blood Test for Traumatic Brain Injury Could Speed Up Treatment for Troops

Posted by in categories: biotech/medical, military, neuroscience

The Food and Drug Administration has approved a blood test to detect concussion that produces results in minutes rather than hours — a breakthrough that could help expedite treatment for service members with traumatic brain injuries, according to the U.S. Army and Abbott Laboratories, the…


Read Next: New Cold-Assignment Incentive Pay Coming for Airmen and Guardians at 7 Bases

“This can help get the most severely injured service members to neurosurgeons faster and ultimately save lives,” Lt. Col. Bradley Dengler, neurosurgical consultant to the U.S. Army Office of the Surgeon General, said in a release.

Continue reading “Newly Approved Rapid Blood Test for Traumatic Brain Injury Could Speed Up Treatment for Troops” »

Apr 5, 2024

Training LLMs over Neurally Compressed Text

Posted by in category: futurism

Google announces Training LLMs over Neurally Compressed Text.

https://huggingface.co/papers/2404.

In this paper, we explore the idea of training large language models (#LLMs) over highly compressed text.

Continue reading “Training LLMs over Neurally Compressed Text” »

Apr 4, 2024

New Latrodectus malware replaces IcedID in network breaches

Posted by in categories: cybercrime/malcode, finance

A relatively new malware called Latrodectus is believed to be an evolution of the IcedID loader, seen in malicious email campaigns since November 2023.

The malware was spotted by researchers at Proofpoint and Team Cymru, who worked together to document its capabilities, which are still unstable and experimental.

IcedID is a malware family first identified in 2017 that was originally classified as a modular banking trojan designed to steal financial information from infected computers. Over time, it became more sophisticated, adding evasion and command execution capabilities.

Apr 4, 2024

The Biggest Takeaways from Recent Malware Attacks

Posted by in categories: cybercrime/malcode, education

Recent high-profile malware attacks teach us lessons on limiting malware risks at organizations. Learn more from Blink Ops about what these attacks taught us.

Apr 4, 2024

New HTTP/2 DoS attack can crash web servers with a single connection

Posted by in category: futurism

Newly discovered HTTP/2 protocol vulnerabilities called “CONTINUATION Flood” can lead to denial of service (DoS) attacks, crashing web servers with a single TCP connection in some implementations.

HTTP/2 is an update to the HTTP protocol standardized in 2015, designed to improve web performance by introducing binary framing for efficient data transmission, multiplexing to allow multiple requests and responses over a single connection, and header compression to reduce overhead.

The new CONTINUATION Flood vulnerabilities were discovered by researcher Barket Nowotarski, who says that it relates to the use of HTTP/2 CONTINUATION frames, which are not properly limited or checked in many implementations of the protocol.

Apr 4, 2024

Microsoft still unsure how hackers stole MSA key in 2023 Exchange attack

Posted by in category: cybercrime/malcode

The U.S. Department of Homeland Security’s Cyber Safety Review Board (CSRB) has released a scathing report on how Microsoft handled its 2023 Exchange Online attack, warning that the company needs to do better at securing data and be more truthful about how threat actors stole an Azure signing key.

Microsoft believes that last May’s Exchange Online hack is linked to a threat actor known as ‘Storm-0558’ stealing an Azure signing key from an engineer’s laptop that was previously compromised by the hackers at an acquired company.

Storm-0558 is a cyberespionage actor affiliated with China that has been active for more than two decades targeting a wide range of organizations.

Apr 4, 2024

Critical flaw in LayerSlider WordPress plugin impacts 1 million sites

Posted by in category: cybercrime/malcode

A premium WordPress plugin named LayerSlider, used in over one million sites, is vulnerable to unauthenticated SQL injection, requiring admins to prioritize applying security updates for the plugin.

LayerSlider is a versatile tool for creating responsive sliders, image galleries, and animations on WordPress sites, allowing users to build visually appealing elements with dynamic content on online platforms.

Researcher AmrAwad discovered the critical (CVSS score: 9.8) flaw, tracked as CVE-2024–2879, on March 25, 2024, and reported it to WordPress security firm Wordfence via its bug bounty program. For his responsible reporting, AmrAwad received a bounty of $5,500.

Apr 4, 2024

Google fixes one more Chrome zero-day exploited at Pwn2Own

Posted by in category: cybercrime/malcode

Google has fixed another zero-day vulnerability in the Chrome browser, which was exploited by security researchers during the Pwn2Own hacking contest last month.

Tracked as CVE-2024–3159, this high-severity security flaw is caused by an out-of-bounds read weakness in the Chrome V8 JavaScript engine.

Remote attackers can exploit the vulnerability using crafted HTML pages to gain access to data beyond the memory buffer via heap corruption, which can provide them with sensitive information or trigger a crash.

Page 203 of 11,135First200201202203204205206207Last