Ai framework finds new dark matter signals by exploring theories
Searching for BSM Experimental Signatures with Large Lagrangian Models
Artificial Intelligence
Summary
Finding signals for new physics beyond the known particles is hard because many theories predict similar signs, and experiments don’t always tell them apart. The authors created hAIthem, a system that uses artificial intelligence to explore many possible models and find areas where existing experiments have not ruled them out. Their AI plays a game to search through complex physics models and then suggests new things to look for in experiments. This approach helps scientists find interesting new directions to test theories about dark matter and other unknown particles.
What this means in practice
- •For particle physics experimental teams: Identify unexplored regions of theoretical models to design new experiments that target previously overlooked dark matter signatures.
- •For machine learning engineers: Use combined reinforcement learning and language model frameworks to optimize search strategies in high-dimensional scientific parameter spaces.
Authors
Ibrahim Elsharkawy, Victoria Knapp-Perez, Wahid Bhimji, Aishik Ghosh
Abstract
The search for physics Beyond the Standard Model (BSM) is generally limited not by the supply of theory descriptions but by the lack of discriminating experimental observations. A case in point is dark matter, where the overwhelming gravitational evidence only goes so far in distinguishing between models within a vast theory space. Exploring the space of testable model signatures may help identify overlooked experimental observables and indicate the utility of future experiments. A challenge is designing a search through model signatures outside what is found in the literature. Our primary contribution is hAIthem, a framework that combines the self-guided exploration of reinforcement learning (RL) with the broad literature-derived knowledge of LLMs. We build an RL agent that learns to find which portions of a theory's high-dimensional parameter space are not excluded under some subset of constraints by playing a Battleship-style "game" against a suite of phenomenology tools. The agent is built as a Large Lagrangian Model (LLaM), an autoregressive transformer that reads a tokenized Lagrangian, is pretrained at scale (here on ~1 billion tokens from ~10,000 Lagrangians), and is fine-tuned in a live environment. The framework then constructs a decision tree that separates RL-found regions using observables computed with established tools, and passes the remaining degenerate regions to a set of LLM agents that compete to produce realistic signatures. In this proof of concept, RL-search outperforms an evolutionary-algorithm baseline, finding more viable regions with greater physical diversity. In a restricted space of single dark scalar multiplet models, we find that hAIthem proposes interesting combinations of previously studied observables, such as the application of a halo-independent kinematic ratio to paleo-detectors.