Mmastodon SciencePhysics first seen 22 h ago, last 9 h ago, peak #1

Astrophysicist says Claude helped expose artefacts in cosmology simulation

Original: This is not beautiful, but it's a very useful analysis which exposes some numerical artefacts in my new simulation, whic

Astrophysicist Franco Vazza says an AI tool helped him uncover numerical artefacts in a new cosmological simulation that he had previously overlooked. He describes the analysis as unattractive but very useful, saying he worked with an LLM trained on a large custom set of instructions. The post is drawing attention for showing AI being used as a genuine debugging aid in scientific research rather than as a novelty.

Why now: Researchers and observers are interested in a concrete case of an LLM catching real technical flaws in scientific code.

Franco VazzaClaude

Open on mastodon →

Evidence

• This is not beautiful, but it's a very useful analysis which exposes some numerical artefacts in my new simulation, which I have previously underlooked. For this I used # AI trained an LLM model in Claude using more than 10,000 tokens. Wait, actually for this I trained Larry,… · franco_vazza · 24

API: https://socialmediatrends-api.osmike.com/v1/trends/721792