LensAgent: A Self Evolving Agent for Autonomous Physical Inference of Sub-galactic Structure
Xiaotang Feng, Zihan Wang, Zilang Shu, Jean-Paul Kneib, Philip Torr
arXiv:2604.03691·astro-ph.GA·Published 2026-04-04
Probing dark matter distribution on sub-galactic scales is essential for testing the Cold Dark Matter ($Λ$CDM) paradigm. Strong gravitational lensing, as one of the most powerful approach by far, provides a direct, purely gravitational probe of these substructures. However, extracting cosmological constraints is severely bottlenecked by the mass-sheet degeneracy (MSD) and the unscalable nature of manual and neural-network modeling. Here, we introduce LensAgent, a pioneering training-free, large language model (LLM)-driven agentic framework for the autonomous physical inference of mass distributions. Operating as an autonomous scientific agent, LensAgent couples high-level logical reasoning with deterministic physical modeling tools, demonstarting successful reconstruction of mass distribution in SLACS Grade A strong lensing systems. This self-evolving architecture enables the robust extraction of sub-galactic substructures at scale, unlocking the cosmological potential of upcoming wide-field surveys such as the Rubin Observatory (LSST) and Euclid.
TopicsAstrophysics & Cosmology
Tagsdark-matter strong-lensing
arXiv categoriesastro-ph.GA, astro-ph.IM
arXiv abstract pagePDF