Regularized Conventions: Equilibrium Computation as a Model of Pragmatic Reasoning
This work addresses the challenge of formalizing pragmatic reasoning in computational linguistics, offering a novel equilibrium-based approach that is incremental in its improvements over prior models.
The paper tackled the problem of modeling pragmatic language understanding by proposing a model where utterances are produced and understood through regularized equilibria of signaling games, resulting in improved or matched predictions compared to existing models across datasets on pragmatic implicatures.
We present a model of pragmatic language understanding, where utterances are produced and understood by searching for regularized equilibria of signaling games. In this model (which we call ReCo, for Regularized Conventions), speakers and listeners search for contextually appropriate utterance--meaning mappings that are both close to game-theoretically optimal conventions and close to a shared, ''default'' semantics. By characterizing pragmatic communication as equilibrium search, we obtain principled sampling algorithms and formal guarantees about the trade-off between communicative success and naturalness. Across several datasets capturing real and idealized human judgments about pragmatic implicatures, ReCo matches or improves upon predictions made by best response and rational speech act models of language understanding.