Hacker News: lupsasca

New comment by lupsasca in "GPT-5.2 derives a new result in theoretical physics"

lupsasca — Sat, 14 Feb 2026 18:30:36 +0000

Hi, I'm an author on the paper. It was definitely a human-AI collaboration, but it is also true that the final simplified formula, Eq. 39 in the paper (which is what we had been seeking, without success), was conjectured and proved by GPT. So it derived a new result in theoretical physics. I'm genuinely puzzled by your complaint.

New comment by lupsasca in "GPT-5.2 derives a new result in theoretical physics"

lupsasca — Sat, 14 Feb 2026 05:52:22 +0000

Correct, on both counts!

New comment by lupsasca in "GPT-5.2 derives a new result in theoretical physics"

lupsasca — Fri, 13 Feb 2026 23:26:11 +0000

My pleasure---thank you for your interest!

New comment by lupsasca in "GPT-5.2 derives a new result in theoretical physics"

lupsasca — Fri, 13 Feb 2026 21:57:50 +0000

There are closely related "MHV amplitudes" which naively obey a really complicated formula, but for which there famously also exists a much simpler "Parke-Taylor formula". Alfredo had derived a complicated expression for these new "single-minus amplitudes" and we were hoping we could find an analogue of the simpler "Parke-Taylor formula" for them.

New comment by lupsasca in "GPT-5.2 derives a new result in theoretical physics"

lupsasca — Fri, 13 Feb 2026 21:43:58 +0000

That paper from the 80s (which is cited in the new one) is about "MHV amplitudes" with two negative-helicity gluons, so "double-minus amplitudes". The main significance of this new paper is to point out that "single-minus amplitudes" which had previously been thought to vanish are actually nontrivial. Moreover, GPT-5.2 Pro computed a simple formula for the single-minus amplitudes that is the analogue of the Parke-Taylor formula for the double-minus "MHV" amplitudes.

New comment by lupsasca in "GPT-5.2 derives a new result in theoretical physics"

lupsasca — Fri, 13 Feb 2026 21:43:23 +0000

Hi I am an author of the paper. We believed that a simple formula should exist but had not been able to find it despite significant effort. It was a collaborative effort but GPT definitely solved the problem for us.

New comment by lupsasca in "Prism"

lupsasca — Tue, 27 Jan 2026 22:06:20 +0000

By "grabbing references" I meant queries of the type "add paper [bla] to the bibliography" -- that seems useful to me!

New comment by lupsasca in "Prism"

lupsasca — Tue, 27 Jan 2026 20:24:55 +0000

I am very sympathetic to your point of view, but let me offer another perspective. First off, you can already vibe-write slop papers with AI, even in LaTeX format--tools like Prism are not needed for that. On the other hand, it can really help researchers improve the quality of their papers. I'm someone who collaborates with many students and postdocs. My time is limited and I spend a lot of it on LaTeX drudgery that can and should be automated away, so I'm excited for Prism to save time on writing, proofreading, making TikZ diagrams, grabbing references, etc.

New comment by lupsasca in "Show HN: Physically accurate black hole simulation using your iPhone camera"

lupsasca — Wed, 20 Nov 2024 02:45:44 +0000

About ~5x improvement. Recall that the resolution of an interferometric array is set by the distance between telescopes measured in units of the observation wavelength. BHEX will get a ~3x resolution improvement from the increased distance between the space satellite and our ground telescopes (for the EHT, the max telescope separation is limited by the diameter of the Earth) and another ~50% from the increased frequency of observations (going up from 230 to 320 GHz).

5x is actually a lot: we'll be able to resolve the "photon ring" of orbiting light around M87* and Sgr A* (the two black holes previously imaged by EHT at lower resolution) and likely see the "shadows" of another 6-8 black holes, with the possibility of estimating the mass of another ~20-30 sources.

New comment by lupsasca in "Show HN: Physically accurate black hole simulation using your iPhone camera"

lupsasca — Wed, 20 Nov 2024 02:40:52 +0000

We're thinking about it! Likely 2025 though.

New comment by lupsasca in "Show HN: Physically accurate black hole simulation using your iPhone camera"

lupsasca — Wed, 20 Nov 2024 00:22:43 +0000

If you put the source infinitely far away (as we are doing here) and at zero velocity relative to the camera, then there is no redshift effect at all, so you could say that this is the choice of redshift that we made :)

If you want the details, they're too long to put in this comment but essentially what I mean is that the r->infty limit of the redshift factor in Eq. (B22) of this paper is unity: https://arxiv.org/abs/2211.07469

New comment by lupsasca in "Show HN: Physically accurate black hole simulation using your iPhone camera"

lupsasca — Wed, 20 Nov 2024 00:20:27 +0000

Not currently. As I mentioned elsewhere, the rotating black hole version that we implemented requires GPU code, and porting that to Android is nontrivial---though we'd love it if someone took our open-source code and ported it!

In the meantime, check out this code developed by Dominic Chang (grad student at Harvard) that implements lensing by a non-rotating (Schwarzschild) black hole in your browser: https://dominic-chang.com/bhi-filter/

New comment by lupsasca in "Show HN: Physically accurate black hole simulation using your iPhone camera"

lupsasca — Wed, 20 Nov 2024 00:19:27 +0000

Excellent! Hope it looked cool.

New comment by lupsasca in "Show HN: Physically accurate black hole simulation using your iPhone camera"

lupsasca — Wed, 20 Nov 2024 00:18:46 +0000

Hahaha, I half-expected a plasma physicist to complain that GRRMHD is not self-consistent and we need to include non-ideal fluid effects :D

New comment by lupsasca in "Show HN: Physically accurate black hole simulation using your iPhone camera"

lupsasca — Wed, 20 Nov 2024 00:17:38 +0000

That would be cool, but then you wouldn't be seeing the world around you anymore. In other words, at that point it becomes a GRMHD simulation, and there is no point in using cameras since the user's environment is obscured, no? Or did you have something else in mind?

New comment by lupsasca in "Show HN: Physically accurate black hole simulation using your iPhone camera"

lupsasca — Wed, 20 Nov 2024 00:15:45 +0000

I haven't looked into the equations but I expect the effect of varying the electric charge would be similar to (but less dramatic than) changing the spin of the black hole, which as you can see in our app is not that big of a change.

New comment by lupsasca in "Show HN: Physically accurate black hole simulation using your iPhone camera"

lupsasca — Wed, 20 Nov 2024 00:14:04 +0000

The rotating black hole version that we implemented requires GPU code, and porting that to Android is nontrivial---though we'd love it if someone took our open-source code and ported it!

New comment by lupsasca in "Show HN: Physically accurate black hole simulation using your iPhone camera"

lupsasca — Wed, 20 Nov 2024 00:02:34 +0000

What the black hole looks like depends on how you define your field of view. And if the black hole is spinning, then you don't expect a circular shadow at all. But in our app, if you pick the "Static black hole" (the non-rotating, Schwarzschild case) and select the "Full FOV" option, then you will see the circular shadow that you expect.

New comment by lupsasca in "Show HN: Physically accurate black hole simulation using your iPhone camera"

lupsasca — Tue, 19 Nov 2024 20:05:23 +0000

Yes, this is one of the wonderful crazy properties of black holes: they get hotter as they evaporate! (More precisely, the Hawking temperature is inversely proportional to the mass!)

New comment by lupsasca in "Show HN: Physically accurate black hole simulation using your iPhone camera"

lupsasca — Tue, 19 Nov 2024 19:18:22 +0000

Glad to hear that! You'll probably also enjoy reading about the Black Hole Explorer (BHEX): a proposed space mission that will take the sharpest images in the history of astronomy and resolve the "photon ring" of orbiting light around a black hole. https://www.blackholeexplorer.org/