Humans often express their communicative intents indirectly or non-literally, which requires their interlocutors—human or AI—to understand beyond the literal meaning of words. While most existing work has focused on discriminative evaluations, we present a new approach to generatively evaluate large language models’ (LLMs’) intention understanding by examining their responses to non-literal utterances. Ideally, an LLM should respond in line with the true intention of a non-literal utterance, not its literal interpretation. Our findings show that LLMs struggle to generate contextually relevant responses to non-literal language. We also find that providing oracle intentions substantially improves response appropriateness, but using chain-of-thought to make models spell out intentions before responding improves much less. These findings suggest that LLMs are not yet pragmatic interlocutors, and that explicitly modeling intention could improve LLM responses to non-literal language.

Large language models (LLMs) excel at processing and generating text and code. However, LLMs have had limited applicability in grounded task-oriented dialogue as they are difficult to steer toward task objectives and fail to handle novel grounding. We present a modular and interpretable grounded dialogue system that addresses these shortcomings by composing LLMs with a symbolic planner and grounded code execution. Our system, consists of a reader and planner: the reader leverages an LLM to convert partner utterances into executable code, calling functions that perform grounding. The translated code’s output is stored to track dialogue state, while a symbolic planner determines the next appropriate response. We evaluate our system’s performance on the demanding OneCommon dialogue task, involving collaborative reference resolution on abstract images of scattered dots. Our system substantially outperforms the previous state-of-the-art, including improving task success in human evaluations from 56% to 69% in the most challenging setting.

Large language models (LLMs) excel at processing and generating both text and code. However, LLMs have had limited applicability in grounded task-oriented dialogue as they are difficult to steer toward task objectives and fail to handle novel grounding. We present a modular and interpretable grounded dialogue system that addresses these shortcomings by composing LLMs with a symbolic planner and grounded code execution. Our system consists of a reader and planner: the reader leverages an LLM to convert partner utterances into executable code, calling functions that perform grounding. The translated code’s output is stored to track dialogue state, while a symbolic planner determines the next appropriate response. We evaluate our system’s performance on the demanding OneCommon dialogue task, involving collaborative reference resolution on abstract images of scattered dots. Our system substantially outperforms the previous state-of-the-art, including improving task success in human evaluations from 56% to 69% in the most challenging setting.

Learning linguistic generalizations from only a few examples is a challenging task. Recent work has shown that program synthesis – a method to learn rules from data in the form of programs in a domain-specific language – can be used to learn phonological rules in highly data-constrained settings. In this paper, we use the problem of phonological stress placement as a case to study how the design of the domain-specific language influences the generalization ability when using the same learning algorithm. We find that encoding the distinction between consonants and vowels results in much better performance, and providing syllable-level information further improves generalization. Program synthesis, thus, provides a way to investigate how access to explicit linguistic information influences what can be learnt from a small number of examples.

Neural models excel at extracting statistical patterns from large amounts of data, but struggle to learn patterns or reason about language from only a few examples. In this paper, we ask: Can we learn explicit rules that generalize well from only a few examples? We explore this question using program synthesis. We develop a synthesis model to learn phonology rules as programs in a domain-specific language. We test the ability of our models to generalize from few training examples using our new dataset of problems from the Linguistics Olympiad, a challenging set of tasks that require strong linguistic reasoning ability. In addition to being highly sample-efficient, our approach generates human-readable programs, and allows control over the generalizability of the learnt programs.