This paper systematically compares different methods of deriving item-level predictions of language models for multiple-choice tasks. It compares scoring methods for answer options based on free generation of responses, various probability-based scores, a Likert-scale style rating method, and embedding similarity. In a case study on pragmatic language interpretation, we find that LLM predictions are not robust under variation of method choice, both within a single LLM and across different LLMs. As this variability entails pronounced researcher degrees of freedom in reporting results, knowledge of the variability is crucial to secure robustness of results and research integrity.
Evaluating grounded neural language model performance with respect to pragmatic qualities like the trade off between truthfulness, contrastivity and overinformativity of generated utterances remains a challenge in absence of data collected from humans. To enable such evaluation, we present a novel open source image-text dataset "Annotated 3D Shapes" (A3DS) comprising over nine million exhaustive natural language annotations and over 12 million variable-granularity captions for the 480,000 images provided by Burges & Kim (2018). We showcase the evaluation of pragmatic abilities developed by a task-neutral image captioner fine-tuned in a multi-agent communication setting to produce contrastive captions. The evaluation is enabled by the dataset because the exhaustive annotations allow to quantify the presence of contrastive features in the model's generations. We show that the model develops human-like patterns (informativity, brevity, over-informativity for specific features (e.g., shape, color biases)).
When faced with a polar question, speakers often provide overinformative answers going beyond a simple "yes" or "no". But what principles guide the selection of additional information? In this paper, we provide experimental evidence from two studies suggesting that overinformativeness in human answering is driven by considerations of relevance to the questioner's goals which they flexibly adjust given the functional context in which the question is uttered. We take these human results as a strong benchmark for investigating question-answering performance in state-of-the-art neural language models, conducting an extensive evaluation on items from human experiments. We find that most models fail to adjust their answering behavior in a human-like way and tend to include irrelevant information. We show that GPT-3 is highly sensitive to the form of the prompt and only achieves human-like answer patterns when guided by an example and cognitively-motivated explanation.
Recent advances in computational cognitive science (i.e., simulation-based probabilistic programs) have paved the way for significant progress in formal, implementable models of pragmatics. Rather than describing a pragmatic reasoning process in prose, these models formalize and implement one, deriving both qualitative and quantitative predictions of human behavior -- predictions that consistently prove correct, demonstrating the viability and value of the framework. The current paper provides a practical introduction to and critical assessment of the Bayesian Rational Speech Act modeling framework, unpacking theoretical foundations, exploring technological innovations, and drawing connections to issues beyond current applications.
While a large body of work has scrutinized the meaning of conditional sentences, considerably less attention has been paid to formal models of their pragmatic use and interpretation. Here, we take a probabilistic approach to pragmatic reasoning about conditionals which flexibly integrates gradient beliefs about richly structured world states. We model listeners' update of their prior beliefs about the causal structure of the world and the joint probabilities of the consequent and antecedent based on assumptions about the speaker's utterance production protocol. We show that, when supplied with natural contextual assumptions, our model uniformly explains a number of inferences attested in the literature, including epistemic inferences, Conditional Perfection and the dependency between antecedent and consequent of a conditional. We argue that this approach also helps explain three puzzles introduced by Douven (2012) about updating with conditionals: depending on the utterance context, the listener's belief in the antecedent may increase, decrease or remain unchanged.
Languages are powerful solutions to coordination problems: they provide stable, shared expectations about how the words we say correspond to the beliefs and intentions in our heads. Yet language use in a variable and non-stationary social environment requires linguistic representations to be flexible: old words acquire new ad hoc or partner-specific meanings on the fly. In this paper, we introduce a hierarchical Bayesian theory of convention formation that aims to reconcile the long-standing tension between these two basic observations. More specifically, we argue that the central computational problem of communication is not simply transmission, as in classical formulations, but learning and adaptation over multiple timescales. Under our account, rapid learning within dyadic interactions allows for coordination on partner-specific common ground, while social conventions are stable priors that have been abstracted away from interactions with multiple partners. We present new empirical data alongside simulations showing how our model provides a cognitive foundation for explaining several phenomena that have posed a challenge for previous accounts: (1) the convergence to more efficient referring expressions across repeated interaction with the same partner, (2) the gradual transfer of partner-specific common ground to novel partners, and (3) the influence of communicative context on which conventions eventually form.