Multilinguality in LLM-designed Reward Functions for RMABs (Collaboration with Google DeepMind)

Chandrasekar Subramanian , Ambreesh Parthasarathy , Ganesh Senrayan , Shreyash Adappanavar , Aparna Taneja , Balaraman Ravindran , Milind Tambe

Technical Research , LLM Agents , Fairness/Bias , Multilingual/Local Language

Description:

Studied multilinguality effects in LLM-designed reward functions for Restless Multi-Armed Bandits (RMABs) on task performance and fairness, with Prof. Milind Tambe’s team.