Pragmatic Language Interpretation as Probabilistic Inference.

By Noah D. Goodman, Michael C. Frank. 2016

665 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Language and Linguistics, Probabilistic & Bayesian Methods, Computational Linguistics & Semantics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Language and Linguistics

In Probabilistic & Bayesian Methods

In Computational Linguistics & Semantics

Citations in the corpus

Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMs (2022, background): “As discussed in §2, pragmatics provides a connection between language development and Theory of Mind (Sperber and Wilson, 1986; Miller, 2006; Tauzin and Gergely, 2018): learning to communicate effectively with language requires reasoning about what our interlocutor knows or does not know (Grice, 197”
Inferring Rewards from Language in Context (2022, background): “A long line of work on pragmatics (Grice, 1975), particularly in the Rational Speech Acts (RSA) framework (Goodman and Frank, 2016), has developed computational models for inferring the behavior or belief that a speaker wishes to induce in a listener.”
Continual Adaptation for Efficient Machine Communication (2019, background): “reasoning about the other agent is typically considered at the time of action selection (i.e. when the speaker is choosing an utterance, or when the listener is choosing a referent; Goodman and Frank, 2016; Andreas and Klein, 2016), this likelihood”
Adversarial Language Games for Advanced Natural Language Intelligence (2019, background): “…reasoning ability of NLP models with pragmatics language games (Krauss and Weinheimer, 1964; Clark and Wilkes-Gibbs, 1986; Potts, 2012) and pragmatics theories such as the speech-act theory (Searle et al., 1980) and Rational Speech Act framework (Golland et al., 2010; Goodman and Frank, 2016).”
Introduction to MAchine Learning & Knowledge Extraction (MAKE) (2017, background): “Moreover, there is empirical evidence that, in some cases, a representation of the speakers’ intentions is helpful [72], and there is agreement that understanding language (not mere language processing) is more than the use of fixed conventions and/or decoding combinatorial structures and that proba”
Interactive Learning from Activity Description (2021, background): “Another related line of research is work on the rational speech act (RSA) or pragmatic reasoning (Grice, 1975; Golland et al., 2010; Monroe & Potts, 2015; Goodman & Frank, 2016; Andreas & Klein, 2016; Fried et al., 2018a), which is also concerned with transferring information via language.”
Semantic Communications for Future Internet: Fundamentals, Applications, and Challenges (2022, neutral): “In linguistics, contextual reasoning is often computationally described using the rational speech act model [118], [119], [120], which is rooted in the Gricean view of language use [121].”
Colors in Context: A Pragmatic Neural Model for Grounded Language Understanding (2017, background): “At its core, the model is a version of the Rational Speech Acts (RSA) model (Frank and Goodman, 2012; Goodman and Frank, 2016), in which agents reason recursively about each other’s expectations and intentions ar X iv :1 70 3.”
Theory of mind as inverse reinforcement learning (2019, background): “, suspecting that the speaker knows there are two cats if we hear them refer to ‘the big cat’ instead of just ‘the cat’) [37]; how we make sense of non-literal word meanings [38]; and even how speakers use prosody to ensure listeners will get the meaning they wish to convey [39] (see [40] for review”
Conscious Processing and the Global Neuronal Workspace Hypothesis (2020, background): “There is currently no good theory of how the brain achieves such meta-representations, although algorithmic models are beginning to be formulated within the abstract framework of lambda calculus (Goodman and Frank, 2016).”
Learning Rewards from Linguistic Feedback (2020, methodology): “While pragmatics are often derived as a result of recursive reasoning (Goodman and Frank 2016), we opt for a simpler heuristic approach.”
Leveraging Language to Learn Program Abstractions and Search Heuristics (2021, background): “However, our formulation also supports T (dt|ρ,L) enriched to include additional priors over language (such as speaker-specific language usage, or pragmatics models that capture a speakers’ other communicative goals (Grice, 1989; Goodman & Frank, 2016).)”