A new neural implant is so small it can rest on a grain of salt, yet it can track and wirelessly transmit brain activity for over a year. It’s powered by laser light that safely passes through tissue and communicates using tiny infrared signals. This ultra-miniature device could transform how scientists study the brain without invasive wiring.
Imagine a world where anyone who needs a bone marrow transplant can get one — on demand. No more desperate donor searches or deadly delays. Kevin Caldwell, Co-Founder & CEO, Ossium Health.
Bone marrow transplants have always depended on finding the right donor at the right time. But what if bone marrow could be stored, shipped, and used on demand—just like a drug? That’s exactly what Ossium Health is now showing in human clinical data.
Kevin Caldwell is the Co-Founder, CEO, and President of Ossium Health (https://ossiumhealth.com/), a clinical-stage bioengineering company pioneering off-the-shelf, cryopreserved bone marrow therapies derived from deceased organ donors.
Under Kevin’s leadership, Ossium has developed a novel platform designed to solve one of the most persistent challenges in transplantation medicine: timely access to compatible bone marrow for patients with life-threatening hematologic malignancies such as Acute Myeloid Leukemia. The company’s approach enables on-demand delivery of viable marrow cells, bypassing the logistical and biological constraints of traditional donor matching and scheduling.
Since its founding, Kevin has scaled Ossium from an early-stage startup into a clinical-stage company with a robust network of over 50 strategic partnerships across supply, clinical development, and commercial channels. He has led multiple financings and secured a landmark contract with the Biomedical Advanced Research and Development Authority, validating Ossium’s relevance to national health preparedness and biomanufacturing resilience.
Dubbed Heretic2R, the sourceport introduces a bunch of new technical features to the elf wizard Corvus’ second adventure. For starters, it adds widescreen support with automatic HUD scaling, and unlocks the framerate to a theoretical maximum of 1,000 FPS.
On top of this, Heretic 2R ensures that in-game special effects update at the appropriate rendering framerate, improves map loading times, and makes “lots of cosmetic improvements” so that “the game plays as you remember it, not as it actually played”. Which sounds very remaster-y to me.
My third-favorite Batman: Arkham game is looking a lot slicker.
Researchers have developed a holographic data storage approach that stores and retrieves information in three dimensions by combining three properties of light—amplitude, phase and polarization. By allowing more data to be stored in the same space, the new approach could help advance efforts to meet the growing global demand for data storage.
Holographic data storage uses laser light to store digital information inside a material. Instead of recording data only on a surface, like a hard drive or optical disk, it stores many overlapping light patterns throughout the volume of the material, allowing much higher storage density and faster data transmission.
“In conventional holographic data storage, data encoding typically uses one light dimension such as amplitude or phase alone, or, at most, combines two of these dimensions,” said research team leader Xiaodi Tan from Fujian Normal University in China.
Tx, Yann LeCun.
• JEPA / H-JEPA: avoids predicting every single pixel (too expensive) and rather predicts in latent space. H-JEPA adds hierarchy — short term details vs long term planning ie. how humans actually learn.
• I-JEPA: built for very efficient vision models. Masks image patches and predicts the semantics and in doing so bypasses heavy compute of traditional autoencoders.
• MC-JEPA & V-JEPA: both of these are built for videos. MC-JEPA separates content (what an object is) vs motion (how it moves). V-JEPA masks video features with no text labels making it perfect of action tracking at scale.
• Audio-JEPA: filters out background noise by treating sounds like visuals.
• Point-JEPA & 3D-JEPA: used primarily in AVs. Uses LiDAR point clouds & volumetric grids.
• ACT-JEPA: filters out real world noise to learn manipulation tasks efficiently via imitation learning.
