This paper provides the first discourse parsing experiments with a large language model (LLM) finetuned on corpora annotated in the style of SDRT (Segmented Discourse Representation Theory, Asher (1993), Asher and Lascarides (2003)). The result is a discourse parser, Llamipa (Llama Incremental Parser), that leverages discourse context, leading to substantial performance gains over approaches that use encoder-only models to provide local, context-sensitive representations of discourse units. Furthermore, it is able to process discourse data incrementally, which is essential for the eventual use of discourse information in downstream tasks.

When engaging in collaborative tasks, humans efficiently exploit the semantic structure of a conversation to optimize verbal and nonverbal interactions. But in recent “language to code” or “language to action” models, this information is lacking. We show how incorporating the prior discourse and nonlinguistic context of a conversation situated in a nonlinguistic environment can improve the “language to action” component of such interactions. We finetune an LLM to predict actions based on prior context; our model, Nebula, doubles the net-action F1 score over the baseline on this task of Jayannavar et al. (2020). We also investigate our model’s ability to construct shapes and understand location descriptions using a synthetic dataset.

We present the MEETING corpus, a dataset of roughly 95 hours of spontaneous meeting-style conversations in French. The corpus is designed to serve as a foundation for downstream tasks such as meeting summarization. In its current state, it offers 25 hours of manually corrected transcripts that are aligned with the audio signal, making it a valuable resource for evaluating ASR and speaker recognition systems. It also includes automatic transcripts and alignments of the whole corpus which can be used for downstream NLP tasks. The aim of this paper is to describe the conception, production and annotation of the corpus up to the transcription level as well as to provide statistics that shed light on the main linguistic features of the corpus.

We provide a new linguistic resource: The Minecraft Structured Dialogue Corpus (MSDC), a discourse annotated version of the Minecraft Dialogue Corpus (MDC; Narayan-Chen et al., 2019), with complete, situated discourse structures in the style of SDRT (Asher and Lascarides, 2003). Our structures feature both linguistic discourse moves and nonlinguistic actions. To show computational tractability, we train a discourse parser with a novel “2 pass architecture” on MSDC that gives excellent results on attachment prediction and relation labeling tasks especially long distance attachments.

In this paper, we present LinTO, an intelligent voice platform and smart room assistant for improving efficiency and productivity in business. Our objective is to build a Spoken Language Understanding system that maintains high performance in both Automatic Speech Recognition (ASR) and Natural Language Processing while being portable and scalable. In this paper we describe the LinTO architecture and our approach to ASR engine training which takes advantage of recent advances in deep learning while guaranteeing high-performance real-time processing. Unlike the existing solutions, the LinTO platform is open source for commercial and non-commercial use

This paper investigates the advantages and limits of data programming for the task of learning discourse structure. The data programming paradigm implemented in the Snorkel framework allows a user to label training data using expert-composed heuristics, which are then transformed via the “generative step” into probability distributions of the class labels given the training candidates. These results are later generalized using a discriminative model. Snorkel’s attractive promise to create a large amount of annotated data from a smaller set of training data by unifying the output of a set of heuristics has yet to be used for computationally difficult tasks, such as that of discourse attachment, in which one must decide where a given discourse unit attaches to other units in a text in order to form a coherent discourse structure. Although approaching this problem using Snorkel requires significant modifications to the structure of the heuristics, we show that weak supervision methods can be more than competitive with classical supervised learning approaches to the attachment problem.

This paper provides a detailed comparison of a data programming approach with (i) off-the-shelf, state-of-the-art deep learning architectures that optimize their representations (BERT) and (ii) handcrafted-feature approaches previously used in the discourse analysis literature. We compare these approaches on the task of learning discourse structure for multi-party dialogue. The data programming paradigm offered by the Snorkel framework allows a user to label training data using expert-composed heuristics, which are then transformed via the “generative step” into probability distributions of the class labels given the data. We show that on our task the generative model outperforms both deep learning architectures as well as more traditional ML approaches when learning discourse structure—it even outperforms the combination of deep learning methods and hand-crafted features. We also implement several strategies for “decoding” our generative model output in order to improve our results. We conclude that weak supervision methods hold great promise as a means for creating and improving data sets for discourse structure.