Corpus ID: 235726325

Builder, we have done it: Evaluating & Extending Dialogue-AMR NLU Pipeline for Two Collaborative Domains

  title={Builder, we have done it: Evaluating \& Extending Dialogue-AMR NLU Pipeline for Two Collaborative Domains},
  author={Claire Bonial and Mitchell Abrams and David R. Traum and Clare R. Voss},
We adopt, evaluate, and improve upon a twostep natural language understanding (NLU) pipeline that incrementally tames the variation of unconstrained natural language input and maps to executable robot behaviors. The pipeline first leverages Abstract Meaning Representation (AMR) parsing to capture the propositional content of the utterance, and second converts this into “Dialogue-AMR,” which augments standard AMR with information on tense, aspect, and speech acts. Several alternative approaches… Expand

Figures and Tables from this paper


Dialogue-AMR: Abstract Meaning Representation for Dialogue
A schema that enriches Abstract Meaning Representation (AMR) in order to provide a semantic representation for facilitating Natural Language Understanding (NLU) in dialogue systems is described and an enhanced AMR that represents not only the content of an utterance, but the illocutionary force behind it, as well as tense and aspect is presented. Expand
Augmenting Abstract Meaning Representation for Human-Robot Dialogue
The design scheme presented here, though task-specific, is extendable for broad coverage of speech acts using AMR in future task-independent work. Expand
Graph-to-Graph Meaning Representation Transformations for Human-Robot Dialogue
A two-step NLU approach is established in which automatically-obtained AMR graphs of the input language are converted into in-domain meaning representation graphs augmented with tense, aspect, and speech act information, thereby bridging the gap from unconstrained natural language input to a fixed set of robot actions. Expand
ISO 24617-2: A semantically-based standard for dialogue annotation
The design and implementation of an incremental method for dialogue act recognition, which proves the usability of the ISO standard for automatic dialogue annotation, is discussed. Expand
Collaborative Dialogue in Minecraft
A Minecraft-based collaborative building task in which one player is shown a target structure and needs to instruct the other player to build this structure, and the subtask of Architect utterance generation is considered, and how challenging it is is considered. Expand
HuRIC: a Human Robot Interaction Corpus
The Human Robot Interaction Corpus (HuRIC) is made of audio files paired with their transcriptions referring to commands for a robot, e.g. in a home environment, to adopt a simple but expressive representation of commands that can be easily translated into the internal representation of the robot. Expand
Toward Abstractive Summarization Using Semantic Representations
This work focuses on the graph-tograph transformation that reduces the source semantic graph into a summary graph, making use of an existing AMR parser and assuming the eventual availability of an AMR-totext generator. Expand
SemBleu: A Robust Metric for AMR Parsing Evaluation
SEMBLEU is a robust metric that extends BLEU to AMRs and punishes situations where a system’s output does not preserve most information from the input, and has slightly higher consistency with human judgments than SMATCH. Expand
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
A new language representation model, BERT, designed to pre-train deep bidirectional representations from unlabeled text by jointly conditioning on both left and right context in all layers, which can be fine-tuned with just one additional output layer to create state-of-the-art models for a wide range of tasks. Expand
Annotation of Tense and Aspect Semantics for Sentential AMR
The proposed framework augments the representation of finite predications to include a four-way temporal distinction and several aspectual distinctions that will enable AMR to be used for NLP tasks and applications that require sophisticated reasoning about time and event structure. Expand