OmniLearned: physics foundation model trained on 1B+ jets hits SOTA on three collider tasks
burny_tech · x · 2026-10-25
A Phys. Rev. D paper presents OmniLearned, a major upgrade of the OmniLearn foundation model for jet physics, with three new elements: updated architecture and training, over one billion training jets, and well-documented software for all datasets and models. Validated on top-quark tagging (Delphes benchmark), b-tagging (ATLAS full simulation), and anomaly detection (CMS experimental data), it achieves state of the art on all three, expanding discovery potential for past, current and future collider experiments.
More from AGI Musings
- Ex-OpenAI researcher: the 'attacker diffusion lag' shielding us from AI misuse may close fast — joshua_saxe · 2026-10-25
- Samotsvety forecasters put max unemployment by 2030 at ~9.7% aggregate (7-9% typical) — NathanpmYoung · 2026-10-25
- Median Americans Think the Typical Corporation Makes 36% Net Profit — NathanpmYoung · 2026-10-25
- Sam Altman: 'I no longer think we're gonna be able to just patch all the bugs' — GarrisonLovely · 2026-10-25
- Finance Practitioner Defends Role-Play Prompts: Benchmark Saturation Isn't ASI — nirmal_dist · 2026-10-25
- Max AI-pilled dev: letting agents work made me treasure real-life time more — adityaag · 2026-10-25