Steering the Herd: A Framework for LLM-based Control of Social Learning

ICLROral2026

Authors: Raghu Arghal, Kevin He, Shirin Saeedi Bidokhti, Saswati Sarkar
Affiliation: University of Pennsylvania, University of Pennsylvania
Venue: ICLR 2026
Track: Oral

TL;DR

We introduce, analyze, and simulate (via LLMs) a model of controlled social learning to study how algorithms can influence social beliefs, decisions, and welfare via information design.

Opening excerpt from the authors’ abstract. source

Read the paper

OpenReview

Topics

control llm

← All ICLR 2026 Oral papers · Browse the whole archive