science
Trending UpSnapshot Window: 2026-10-01 16:10 UTC · ← Back to Non-Crypto Overview
Topic Engagement Summary
Total Engagement - Comments: 62, Retweets: 261, Likes: 1294, Impressions: 58846
Verbatim Citations & Public Discussions 11 source posts analyzed
· 2026-10-01T15:49:36+00:00LIFTOFF: SpaceX sends four astronauts soaring toward the International Space Station as Crew-13 begins its long-duration mission in orbit. NASA astronauts Jessica Watkins and Luke Delaney are joined by Canadian astronaut Joshua Kutryk and Roscosmos cosmonaut Sergey Teteryatnikov
· 2026-10-01T15:19:38+00:00B1101-3 lands at LZ-40.
· 2026-10-01T13:08:46+00:00temperature difference
· 2026-10-01T13:00:53+00:00Getting a live demo of untwisting DNA using molecular nodes from the great @bradyajohnston #b3d #geometrynodes
· 2026-10-01T16:16:25+00:00The Emirates Mars Mission's Hope probe captured its historic first image of Mars from an altitude of 24,700 kilometers in 2021, revealing the towering volcano Olympus Mons, three major shield volcanoes, and the vast canyon system Valles Marineris bathed in early morning sunlight.
AI visual note: A spherical view of Mars dominates the black background, showcasing its rust-colored terrain, dark surface features, and wispy white clouds, captured by the UAE's Hope probe in 2021. The image highlights Olympus Mons and the Valles Marineris canyon system bathed in early morning sunlight from an altitude of 24,700 kilometers.
· 2026-10-01T15:05:19+00:00Reconstructing the Dynamic World: A Representation-Centric View of 4D Scene Reconstruction Ziren Gong, Guo Chen, Yongjia Li, Yihua Shao, @fabiotosi92 , @s_matt , @mattpoggi , @HaoTang_ai , Fei Ma, Shuyan Li, Ziyang Yan, Nicu Sebe, Ling Shao, Jianfei Cai, Qi Tian, @MingHsuanYang
AI visual note: Academic figure showing "Trends in 4D reconstruction," a timeline of publications across Classical, Learning-based, NeRFs and Implicit Radiance Fields, and Gaussian Splatting categories from pre-2015 to present.
· 2026-10-01T15:39:33+00:00The SpaceX account will end up being a 24/7 live video coverage of constant starship launches.
· 2026-10-01T15:52:00+00:00The Andromeda Galaxy!
· 2026-10-01T15:21:28+00:00Godspeed crew 13, thank you for flying spacex
· 2026-10-01T14:36:46+00:00Check out our new work on continual learning of dynamical systems! CRUG lets a single RNN learn new systems while preserving earlier dynamics, allocating units as needed and recycling unused capacity. arxiv.org/abs/2609.38356
AI visual note: The image is a figure for a research paper titled "Learn new dynamical systems without forgetting the previous ones," introducing a method called **CRUG (Continental Learning in RNNs through Recyclable Unit Gating)**. It illustrates a three-step process where multiple dynamical systems (task 1, task 2, task 3, represented as distinct waveforms like a butterfly-shaped attractor and spiraling patterns) arrive sequentially over time; during training, a pool of RNN units is opened, compactly allocated to each task, and then unused units are recycled (marked as free or non-readout), while a single inference network uses one readout group per task to reconstruct all the learned systems without forgetting.
· 2026-10-01T14:48:18+00:00Good night everyone It is a law of nature that night falls as the sun sets in the western sky at the end of the day. We humans work hard throughout the day, leaving our bodies exhausted; consequently, we sleep to replenish our energy for the next day. Sleep is beneficial for
AI visual note: A peaceful nighttime illustration shows a young man sleeping peacefully in bed beside a window with a moonlit city view, accompanied by "Good Night" greetings, "SANSEM Believe Focus Achieve" signage, and "Sweet Dreams!" text. The image visually complements the post's reflection on the natural cycle of night falling and the restorative power of sleep after a hard day's work.