🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Paper day! arxiv.org/abs/2412.05806 I took some time to revisit statistical theory and ended up writing a paper about something that's been on my mind. The work explores why machine learning predictions in astronomy often don't perfectly align with the ideal 1:1 line. While

NeurIPS acknowledges that the cultural generalization made by the keynote speaker today reinforces implicit biases by making generalisations about Chinese scholars. This is not what NeurIPS stands for. NeurIPS is dedicated to being a safe space for all of us. We want to address

A huge thanks to Lindsay Oldham and Nature Astronomy for featuring our AstroMLab research in their latest highlights! It's been challenging to find the right home for our papers that bridge LLM and astronomy, so I'm thrilled to see astronomy journals becoming more welcoming of

Finally got those 3Blue1Brown-style Manim animations done for today's lecture.

Paper day: led by Erwin Chen Erwin Chen Using Gaia XP spectra to probe very metal-poor stars, we discovered a distinct population concentrated in the inner Galaxy, manifesting as an overdensity in the metallicity distribution around [Fe/H] = -2.3. This reveals a massive

Paper day! Led by Jiadong Li Jiadong Li Double kill with Gaia XP spectra of the day! We showed that even with XP spectra's low resolution, we could perform a mixture of white dwarf and main sequence composite spectra through neural network emulators to find a vast number of

I recently contributed a piece to Nature Astronomy on how large language models might reshape astronomy researcher identity and education - a topic I've been thinking about quite a bit lately. The piece is available here: rdcu.be/ebFws

A true loss to the community. Bob embodied one of the most admirable traits I have been wanting to emulate: focus, diligence, and indifference to optics and fame. Speaking with him made me realize how long a road I still had ahead to become a great scientist like him.

Paper day: Led by the one and only Tomasz Różański, we conducted a painstakingly detailed study on scaling spectral emulations with Transformers. We learned how to optimally predict the number of training samples needed in balance with model size and hyperparameter choices. The

Can't help myself.

Tidying my lecture notes from this semester. This came to mind.

Awesome paper from Yuan-Sen Ting 丁源森 on Teaching Astronomy with LLMs "Our findings suggest that structured LLM integration with transparency requirements and domain-specific tools can enhance astronomy education while building essential AI literacy skills." arxiv.org/abs/2506.06921