Zero-Shot Transfer Learning with Synthesized Data for Multi-Domain Dialogue State Tracking

@inproceedings{Campagna2020ZeroShotTL,
  title={Zero-Shot Transfer Learning with Synthesized Data for Multi-Domain Dialogue State Tracking},
  author={Giovanni Campagna and Agata Foryciarz and M. Moradshahi and Monica S. Lam},
  booktitle={ACL},
  year={2020}
}
Zero-shot transfer learning for multi-domain dialogue state tracking can allow us to handle new domains without incurring the high cost of data acquisition. This paper proposes new zero-short transfer learning technique for dialogue state tracking where the in-domain training data are all synthesized from an abstract dialogue model and the ontology of the domain. We show that data augmentation through synthesized data can improve the accuracy of zero-shot learning for both the TRADE model and… 

Figures and Tables from this paper

Zero-Shot Dialogue State Tracking via Cross-Task Transfer
TLDR
This work proposes TransferQA, a transferable generative QA model that seamlessly combines extractive QA and multi-choice QA via a text-to-text transformer framework, and tracks both categorical slots and non-categorical slots in DST.
Leveraging Slot Descriptions for Zero-Shot Cross-Domain Dialogue StateTracking
TLDR
This model first encodes a dialogue context and a slot with a pre-trained self-attentive encoder, and generates slot value in auto-regressive manner, and incorporates Slot Type Informed Descriptions that capture the shared information of different slots to facilitates the cross-domain knowledge transfer.
Few Shot Dialogue State Tracking using Meta-learning
TLDR
This work proposes a meta-learner D-REPTILE specific to the DST problem, which is agnostic of the underlying model and hence any existing state-of-the-art DST system can improve its performance on unknown domains using the training strategy.
Pre-training and Data Augmentation for Dialogue State Tracking
TLDR
The detailed experiments show that baseline models benefit immensely when pre-trained with span-level objectives in multiple phases and suggest that the use of data augmentation techniques such as paraphrasing also improve the performance on DST.
In-Context Learning for Few-Shot Dialogue State Tracking
TLDR
This work proposes an in-context (IC) learning framework for zero-shot and few-shot learning dialogue state tracking (DST), where a large pretrained language model (LM) takes a test instance and a few exemplars as input, and directly decodes the dialogue state without any parameter updates.
N-Shot Learning for Augmenting Task-Oriented Dialogue State Tracking
TLDR
An augmentation framework that utilizes belief state annotations to match turns from various dialogues and form new synthetic dialogues in a bottom-up manner is introduced, concluding that exploiting belief state Annotations enhances dialogue augmentation and results in improved models in n-shot training scenarios.
Prompt Learning for Few-Shot Dialogue State Tracking
TLDR
A prompt learning framework for few-shot DST, which consists of two main components: value-based prompt and inverse prompt mechanism, that indicates that DST-related knowledge can be probed from PLM and utilized to address low-resource DST efficiently with the help of prompt learning.
Improving Dialogue State Tracking with Turn-based Loss Function and Sequential Data Augmentation
TLDR
A Turn-based Loss Function (TLF) that penalises the model if it inaccurately predicts a slot value at the early turns more so than in later turns in order to improve joint goal accuracy and a simple but effective Sequential Data Augmentation (SDA) algorithm to generate more complex user utterances and system responses to effectively train existing DST models.
Robust Dialogue State Tracking with Weak Supervision and Sparse Data
TLDR
A training strategy to build extractive DST models without the need for fine-grained manual span labels, and a new model architecture with a unified encoder that supports value as well as slot independence by leveraging the attention mechanism.
NeuralWOZ: Learning to Collect Task-Oriented Dialogue via Model-Based Simulation
TLDR
The proposed NeuralWOZ, a novel dialogue collection framework that uses model-based dialogue simulation, achieves a new state-of-the-art with improvements of 4.4% point joint goal accuracy on average across domains, and improvements of 5.7% point of zero-shot coverage against the MultiWOz 2.1 dataset.
...
...

References

SHOWING 1-10 OF 39 REFERENCES
Transferable Multi-Domain State Generator for Task-Oriented Dialogue Systems
TLDR
A Transferable Dialogue State Generator (TRADE) that generates dialogue states from utterances using copy mechanism, facilitating transfer when predicting (domain, slot, value) triplets not encountered during training.
Zero-Shot Dialog Generation with Cross-Domain Latent Actions
TLDR
Experimental results show that the proposed zero-shot dialog generation method is able to achieve superior performance in learning dialog models that can rapidly adapt their behavior to new domains and suggests promising future research.
MultiWOZ 2.1: Multi-Domain Dialogue State Corrections and State Tracking Baselines
TLDR
This work uses crowdsourced workers to fix the state annotations and utterances in the original version of the MultiWOZ data, hoping that this dataset resource will allow for more effective dialogue state tracking models to be built in the future.
MultiWOZ 2.1: A Consolidated Multi-Domain Dialogue Dataset with State Corrections and State Tracking Baselines
TLDR
This work uses crowdsourced workers to re-annotate state and utterances based on the original utterances in the dataset, and benchmark a number of state-of-the-art dialogue state tracking models on the MultiWOZ 2.1 dataset and show the joint state tracking performance on the corrected state annotations.
An End-to-end Approach for Handling Unknown Slot Values in Dialogue State Tracking
TLDR
An E2E architecture based on the pointer network (PtrNet) that can effectively extract unknown slot values while still obtains state-of-the-art accuracy on the standard DSTC2 benchmark is described.
Toward Scalable Neural Dialogue State Tracking Model
TLDR
A new scalable and accurate neural dialogue state tracking model, based on the recently proposed Global-Local Self-Attention encoder (GLAD) model, which reduces the latency in training and inference times by $35 on average, while preserving performance of belief state tracking.
Multi-domain Dialogue State Tracking as Dynamic Knowledge Graph Enhanced Question Answering
TLDR
This paper proposes to model multi-domain dialogue state tracking as a question answering problem, referred to as Dialogue State Tracking via Question Answering (DSTQA), and uses a dynamically-evolving knowledge graph to explicitly learn relationships between (domain, slot) pairs.
Global-Locally Self-Attentive Encoder for Dialogue State Tracking
TLDR
This paper proposes the Global-Locally Self-Attentive Dialogue State Tracker (GLAD), which learns representations of the user utterance and previous system actions with global-local modules and shows that this significantly improves tracking of rare states.
Neural Belief Tracker: Data-Driven Dialogue State Tracking
TLDR
This work proposes a novel Neural Belief Tracking (NBT) framework which overcomes past limitations, matching the performance of state-of-the-art models which rely on hand-crafted semantic lexicons and outperforming them when such lexicons are not provided.
A Network-based End-to-End Trainable Task-oriented Dialogue System
TLDR
This work introduces a neural network-based text-in, text-out end-to-end trainable goal-oriented dialogue system along with a new way of collecting dialogue data based on a novel pipe-lined Wizard-of-Oz framework that can converse with human subjects naturally whilst helping them to accomplish tasks in a restaurant search domain.
...
...