Related papers: Generalizing meanings from partners to populations: Hierarchical inference supports convention formation on networks

Generalizing meanings from partners to populations: Hierarchical inference supports convention formation on networks

URL: http://arxiv.org/abs/2002.01510v2
Date: Sat, 30 May 2020 07:07:50 GMT
Title: Generalizing meanings from partners to populations: Hierarchical inference supports convention formation on networks
Authors: Robert D. Hawkins, Noah D. Goodman, Adele E. Goldberg, Thomas L. Griffiths
Abstract summary: A key property of linguistic conventions is that they hold over an entire community of speakers. We propose a hierarchical Bayesian model to explain how speakers and listeners solve this inductive problem.
Score: 31.07078356126945
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Abstract: A key property of linguistic conventions is that they hold over an entire community of speakers, allowing us to communicate efficiently even with people we have never met before. At the same time, much of our language use is partner-specific: we know that words may be understood differently by different people based on our shared history. This poses a challenge for accounts of convention formation. Exactly how do agents make the inferential leap to community-wide expectations while maintaining partner-specific knowledge? We propose a hierarchical Bayesian model to explain how speakers and listeners solve this inductive problem. To evaluate our model's predictions, we conducted an experiment where participants played an extended natural-language communication game with different partners in a small community. We examine several measures of generalization and find key signatures of both partner-specificity and community convergence that distinguish our model from alternatives. These results suggest that partner-specificity is not only compatible with the formation of community-wide conventions, but may facilitate it when coupled with a powerful inductive mechanism.

Related papers

Decentralised Emergence of Robust and Adaptive Linguistic Conventions in Populations of Autonomous Agents Grounded in Continuous Worlds [4.63732827131233]
This paper introduces a methodology through which a population of autonomous agents can establish a linguistic convention. The convention emerges in a decentralised manner through local communicative interactions between pairs of agents. We show that the methodology enables a population to converge on a communicatively effective, coherent and human-interpretable linguistic convention.
arXiv Detail & Related papers (2024-01-16T16:11:35Z)
Regularized Conventions: Equilibrium Computation as a Model of Pragmatic Reasoning [72.21876989058858]
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 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.
arXiv Detail & Related papers (2023-11-16T09:42:36Z)
MindDial: Belief Dynamics Tracking with Theory-of-Mind Modeling for Situated Neural Dialogue Generation [62.44907105496227]
MindDial is a novel conversational framework that can generate situated free-form responses with theory-of-mind modeling. We introduce an explicit mind module that can track the speaker's belief and the speaker's prediction of the listener's belief. Our framework is applied to both prompting and fine-tuning-based models, and is evaluated across scenarios involving both common ground alignment and negotiation.
arXiv Detail & Related papers (2023-06-27T07:24:32Z)
Analyzing Gender Representation in Multilingual Models [59.21915055702203]
We focus on the representation of gender distinctions as a practical case study. We examine the extent to which the gender concept is encoded in shared subspaces across different languages.
arXiv Detail & Related papers (2022-04-20T00:13:01Z)
Few-shot Language Coordination by Modeling Theory of Mind [95.54446989205117]
We study the task of few-shot $textitlanguage coordination$. We require the lead agent to coordinate with a $textitpopulation$ of agents with different linguistic abilities. This requires the ability to model the partner's beliefs, a vital component of human communication.
arXiv Detail & Related papers (2021-07-12T19:26:11Z)
Extending rational models of communication from beliefs to actions [10.169856458866088]
Speakers communicate to influence their partner's beliefs and shape their actions. We develop three speaker models: a belief-oriented speaker with a purely informative objective; an action-oriented speaker with an instrumental objective; and a combined speaker which integrates the two. We show that grounding production choices in future listener actions results in relevance effects and flexible uses of nonliteral language.
arXiv Detail & Related papers (2021-05-25T13:58:01Z)
From partners to populations: A hierarchical Bayesian account of coordination and convention [25.131987884154054]
We argue that the central computational problem of communication is not simply transmission, but learning and adaptation over multiple timescales. We present new empirical data alongside simulations showing how our model provides a cognitive foundation for explaining several phenomena.
arXiv Detail & Related papers (2021-04-12T23:00:40Z)
On the Critical Role of Conventions in Adaptive Human-AI Collaboration [73.21967490610142]
We propose a learning framework that teases apart rule-dependent representation from convention-dependent representation. We experimentally validate our approach on three collaborative tasks varying in complexity.
arXiv Detail & Related papers (2021-04-07T02:46:19Z)
Characterizing English Variation across Social Media Communities with BERT [9.98785450861229]
We analyze two months of English comments in 474 Reddit communities. The specificity of different sense clusters to a community, combined with the specificity of a community's unique word types, is used to identify cases where a social group's language deviates from the norm. We find that communities with highly distinctive language are medium-sized, and their loyal and highly engaged users interact in dense networks.
arXiv Detail & Related papers (2021-02-12T23:50:57Z)

This list is automatically generated from the titles and abstracts of the papers in this site.