2 papers
cs.CL2026
LLMs struggle to simulate human belief updates in controlled environments
Sebastian Pohl, Harsh Mehta, Pranav Mambayil +4
LLMs are increasingly deployed as proxies for human study participants in social science experiments, yet the fidelity of this practice has rarely been tested directly. We test whe…
cs.GT2026
Characterisation of reactive Nash equilibria in repeated additive games
Franziska Lesigang, Christian Hilbe, Nikoleta E. Glynatsi
In this paper, we study reactive strategies in repeated additive games between two players with finitely many actions. Reactive strategies condition only on the opponent's previous…