Semantic Parsing for Task Oriented Dialog using Hierarchical Representations
By Sonal Gupta, Rushin Shah, Mrinal Mohit, Anuj Kumar, Michael Lewis. 2018
213 citations recorded by the source corpus.
Task oriented dialog systems typically first parse user utterances to semantic frames comprised of intents and slots. Previous work on task oriented intent and slot-filling work has been restricted to one intent per query and one slot label per token, and thus cannot model complex compositional requests. Alternative semantic parsing systems have represented queries as logical forms, but these are challenging to annotate and parse. We propose a hierarchical annotation scheme for semantic parsing that allows the representation of compositional queries, and can be efficiently and accurately parsed by standard constituency parsing models. We release a dataset of 44k annotated queries (http://fb.me/semanticparsingdialog), and show that parsing models outperform sequence-to-sequence approaches on this dataset.
Subfields: Artificial Intelligence, Dialogue & Conversational AI, Information Extraction
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
- Mohit Bansal
- Luke S. Zettlemoyer
- Percy Liang
- Jianfeng Gao
- Caiming Xiong
- Noah A. Smith
- Richard Socher
- Yejin Choi
In Dialogue & Conversational AI
In Information Extraction
- Daniel Klein
- Percy Liang
- Jacob Andreas
- Dan Roth
- Siva Reddy
- Jayant Krishnamurthy
- Jianpeng Cheng
- Mirella Lapata
Citations in the corpus
- ExT5: Towards Extreme Multi-Task Scaling for Transfer Learning (2021, neutral): “…et al., 2019), DART (Radev et al., 2020), ToTTo (Parikh et al., 2020), Wiki-Auto (Jiang et al., 2020), TurkCorpus(Xu et al., 2016) – DialoGLUE: Banking77 (Casanueva et al., 2020), HWU64 (Liu et al., 2019b), CLINC150 (Larson et al., 2019), SGD (Rastogi et al., 2020), TOP (Gupta et al., 2018).”
- Prompt Conditioned VAE: Enhancing Generative Replay for Lifelong Learning in Task-Oriented Dialogue (2022, methodology): “…slot filling based on public NLU benchmarks: For intent detection, we collect six datasets that carry intent annotations: HWU (Liu et al., 2019), BANKING (Casanueva et al., 2020), CLINC (Lar-son et al., 2019), SNIPS (Coucke et al., 2018), AITS (Hemphill et al., 1990), and TOP (Gupta et al., 2018).”
- Snorkel DryBell: A Case Study in Deploying Weak Supervision at Industrial Scale (2018, background): “We observe other organizations already developing broad-purpose models for domains such as language modeling [13, 14], object recognition [20], and object detection [39].”
- PyText: A Seamless Path from NLP research to production (2018, methodology): “PyText provides a reference implementation for Recurrent Neural Network Grammars (Dyer et al., 2016) (Gupta et al., 2018) for semantic parsing.”
- SeqZero: Few-shot Compositional Semantic Parsing with Sequential Prompts and Zero-shot Models (2022, methodology): “On EcommerceQuery dataset, we use only LSTM seq2seq and BART as baselines, because Iyer et al. (2017) requires user feedbacks, and Zheng and Lapata (2020) requires domain specific semantic tags, which are not available in Ecom-merceQuery.”
- Update Frequently, Update Fast: Retraining Semantic Parsing Systems in a Fraction of Time (2021, methodology): “For example, the publicly available TOP dataset (Gupta et al., 2018) uses 25 intents and 36 slots for just two domains navigation and events .”
- MTOP: A Comprehensive Multilingual Task-Oriented Semantic Parsing Benchmark (2020, methodology): “With the goal of handling more complex queries, recent approaches propose hierarchical representations (Gupta et al., 2018) that are expressive enough to capture the task-specific semantics of complex nested queries.”
- MASSIVE: A 1M-Example Multilingual Natural Language Understanding Dataset with 51 Typologically-Diverse Languages (2022, unknown): “, 2019) 2 4,818 2 14-15 11-12 Task Oriented Parsing (TOP) (Gupta et al., 2018) 1 44,873 2 25 36 Multilingual Task-Oriented Semantic Parsing (MTOP) (Li et al.”
- Controllable Semantic Parsing via Retrieval Augmentation (2021, methodology): “Data The MTOP dataset uses the decoupled TOP representation (Gupta et al., 2018; Li et al., 2021a), as exemplified in Figures 1, 2, and 3.”
- A Closer Look At Feature Space Data Augmentation For Few-Shot Intent Classification (2019, background): “For text classification, several text generation based data augmentation techniques have also been explored (Hou et al., 2018; Zhao et al., 2019; Guu et al., 2018; Yoo et al., 2018; Cho et al., 2019a).”
- Zero-Shot Transfer Learning with Synthesized Data for Multi-Domain Dialogue State Tracking (2020, methodology): “Gupta et al. (2018) previously proposed a different annotation scheme for dialogues, using a hierarchical representation scheme, instead of the more typical intent and slot.”
- A Comprehensive Exploration on WikiSQL with Table-Aware Word Contextualization (2019, background): “, 2018) and navigation control (Gupta et al., 2018) via speechbased smart devices.”
- Practical Semantic Parsing for Spoken Language Understanding (2019, methodology): “A similar approach to transform intent/slot data into tree structures has been recently employed by Gupta et al. (2018b).”
- Building a Neural Semantic Parser from a Domain Ontology (2018, background): “Cheng et al. (2017a, 2017b), Yin and Neubig (2018), Gupta et al. (2018), Chen et al. (2018) all employ neural sequence-to-action models to generate structured meaning representations.”
- DialoGLUE: A Natural Language Understanding Benchmark for Task-Oriented Dialogue (2020, background): “TOP (Gupta et al., 2018) is a dataset of 44k utterances wherein each utterance is annotated with a hierarchical semantic representation.”
- MeLL: Large-scale Extensible User Intent Classification for Dialogue Systems with Meta Lifelong Learning (2021, methodology): “TaskDialog-EUICisbuiltfromthreepublicdatasets: Snips[7], TOP semantic parsing [14] and Facebook Multilingual Task Oriented Dataset [41].”
- Don’t Parse, Generate! A Sequence to Sequence Architecture for Task-Oriented Semantic Parsing (2020, methodology): “works [28]. For the TOP dataset, we pick a model based on Recurrent Neural Network Grammars (RNNG) [3], the Shift Reduce Parser. We provide a brief overview of this model as described in Gupta et al. [6] - the parse tree is constructed using a sequence of transitions, or actions. The transitions are”
- Task-Oriented Dialogue as Dataflow Synthesis (2020, background): “But while recent variants support interaction between related slots (Budzianowski et al., 2018) and fixed-depth hierarchies of slots (Gupta et al., 2018), modern slot-filling approaches remain limited in their support for semantic compositionality.”
- BiToD: A Bilingual Multi-Domain Dataset For Task-Oriented Dialogue Modeling (2021, background): “Different from aforemen305 tioned datasets focus on dialogue level modeling, there are many SLU datasets [40, 41, 42, 43] have 306 been proposed for utterance level semantic parsing.”