Reposted by UniverseTBDMike Smith @mjjsmith.com · 30/06/2026Today is the day you (🫵!) get access to 80TB plus of data from over 30 astronomical surveys in one place. 4GB amount of RAM is enough even at Gaia Scale 🔭. Check it out here: huggingface.co/blog/hugging... 011
Reposted by UniverseTBDMike Smith @mjjsmith.com · 02/06/2025New AstroPT models are out 🔭🎉 This time trained with an improved DESI galaxy image dataset. Link here: huggingface.co/Smith42/astr... Check out these new scaling curves! We are still seeing improvement at 800M parameters where before we stalled at 100M. Maybe high quality data is all you need 🤔 141
UniverseTBD @universetbd.org · 18/04/2025📢 New dataset out! We introduce HypoGen💥, a dataset of ~5.5K structured problem–hypothesis pairs (Bit–Flip–Spark + Chain‑of‑Reasoning) to advance LLM-driven scientific ideation💡. Fine‑tuned LLaMA 3.1 8B & R1‑distilled models show significant gains. Humans are still the best🥇. 154
UniverseTBD @universetbd.org · 16/04/2025Excited to announce our new paper as part of our 2^2 week: AstroLLaVA, a vision language model for astronomy that enables natural dialogue with astronomical imagery! Shout out to Sharaf Zaman for leading this work arxiv.org/abs/2504.08583 🔭☄️ 173
UniverseTBD @universetbd.org · 15/04/2025Our 2^2 celebration is still in full swing! 🎉 Today we’re launching our latest, must-read survey paper: “A Survey on Hypothesis Generation for Scientific Discovery in the Era of Large Language Models.” Check it out! arxiv.org/pdf/2504.054... 🔭 160
UniverseTBD @universetbd.org · 14/04/2025Our 2^2 celebration continues! 🎉 Today we’re excited to introduce AstroCoder – an AI tool that makes niche astronomy software discoverable, scanning over 2,270+ codebases to deliver technical summaries, installation guides, and code examples. 🚀 Try it now at nolank.ca/astrocoder/nolank.caAstroCoder 190
UniverseTBD @universetbd.org · 13/04/2025🎉 HAPPY BIRTHDAY, UniverseTBD! 🚀 As we turn 2, we’re going 2^2. Launching a new project per day for the next four days. We hope that you all enjoy these works as much as we have enjoyed working on them. Stay tuned for the big reveals! 131
UniverseTBD @universetbd.org · 04/02/2025Join us tomorrow (Feb 5th) at 10:00 UTC for our monthly talk 🚀 🔭! Maya Jablonska (ANU) will be presenting: "UniverseTBD: Democratising Science for Everyone" We will be streaming on Discord (discord.gg/tSKx4h8p?eve...) and on Youtube (youtube.com/@UniverseTBD...)discord.ggJoin the UniverseTBD Discord Server!Check out the UniverseTBD community on Discord – hang out with 162 other members and enjoy free voice and text chat. 040
UniverseTBD @universetbd.org · 13/12/2024Such a cool project! Excited to see how Herakoi can integrate with astro LLMs 😼 120
Reposted by UniverseTBDLuca Di Mascolo 🟥 @lucadimascolo.bsky.social · 13/12/2024Tiny, but significant milestone for #herakoi! Earlier this year, the SAIt invited me and @micginolfi.bsky.social to present our "motion-sensing sonification experiment" at the VITEII workshop. And today, we got a short proceeding out on #arxiv: arxiv.org/abs/2412.09152 🌌🔭🐡🧪 163
Reposted by UniverseTBDMike Smith @mjjsmith.com · 04/12/2024Aaand the Multimodal Universe is on ArXiv 🎉 arxiv.org/abs/2412.02527 Come find the poster at #NeurIPS #NeurIPS2024 #astrocode 🔭arxiv.orgThe Multimodal Universe: Enabling Large-Scale Machine Learning with 100TB of Astronomical Scientific DataWe present the MULTIMODAL UNIVERSE, a large-scale multimodal dataset of scientific astronomical data, compiled specifically to facilitate machine learning research. Overall, the MULTIMODAL UNIVERSE co... 0135
UniverseTBD @universetbd.org · 02/12/2024Excited to announce the Multimodal Universe (MMU): a huge 100TB dataset bringing together the largest ML-focused collection of astronomical observations ever assembled to accelerate open AI and astronomy research github.com/MultimodalUn... Think ImageNet, but for space 🔭 #astrocode 🧵 2186
UniverseTBD @universetbd.org · 28/11/2024Very nice work on chemistry and materials science foundation models from Nawaf Alampara and team (including our own @pktrpl.bsky.social 🥳🥳 ) 🧪🧪 Paper link: arxiv.org/abs/2411.16955 031
UniverseTBD @universetbd.org · 27/11/2024Join us on Discord tomorrow (Nov 28th) at 15:00 UTC for our monthly talk 🔭 🧪 ! @micginolfi.bsky.social (U Florence) will be talking about how we can infer many galaxy properties via a single neural net (nice X thread here with more details: x.com/micginolfi/s...). Link: discord.gg/9PXcTH7cTn?e... 132