Transferable Multi-Domain State Generator for Task-Oriented Dialogue Systems
- Chien-Sheng Wu, Andrea Madotto, Ehsan Hosseini-Asl, Caiming Xiong, R. Socher, Pascale Fung
- Computer ScienceAnnual Meeting of the Association for…
- 21 May 2019
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.
A Simple Language Model for Task-Oriented Dialogue
- Ehsan Hosseini-Asl, Bryan McCann, Chien-Sheng Wu, Semih Yavuz, R. Socher
- Computer ScienceNeural Information Processing Systems
- 2 May 2020
SimpleTOD is a simple approach to task-oriented dialogue that uses a single causal language model trained on all sub-tasks recast as a single sequence prediction problem, which allows it to fully leverage transfer learning from pre-trained, open domain, causal language models such as GPT-2.
TOD-BERT: Pre-trained Natural Language Understanding for Task-Oriented Dialogue
- Chien-Sheng Wu, S. Hoi, R. Socher, Caiming Xiong
- Computer ScienceConference on Empirical Methods in Natural…
- 15 April 2020
The experimental results show that the pre-trained task- oriented dialogue BERT (ToD-BERT) surpasses BERT and other strong baselines in four downstream task-oriented dialogue applications, including intention detection, dialogue state tracking, dialogue act prediction, and response selection.
Global-to-local Memory Pointer Networks for Task-Oriented Dialogue
- Chien-Sheng Wu, R. Socher, Caiming Xiong
- Computer ScienceInternational Conference on Learning…
- 15 January 2019
The proposed global-to-local memory pointer networks can improve copy accuracy and mitigate the common out-of-vocabulary problem, and is able to improve over the previous state- of-the-art models in both simulated bAbI Dialogue dataset and human-human Stanford Multi-domain Dialogue dataset on automatic and human evaluation.
Find or Classify? Dual Strategy for Slot-Value Predictions on Multi-Domain Dialog State Tracking
- Jianguo Zhang, Kazuma Hashimoto, Caiming Xiong
- Computer ScienceSTARSEM
- 8 October 2019
This paper proposes a simple yet effective dual-strategy model for DST, by adapting a single BERT-style reading comprehension model to jointly handle both the categorical and non-categorical slots.
Mem2Seq: Effectively Incorporating Knowledge Bases into End-to-End Task-Oriented Dialog Systems
- Andrea Madotto, Chien-Sheng Wu, Pascale Fung
- Computer ScienceAnnual Meeting of the Association for…
- 23 April 2018
This paper empirically shows how Mem2Seq controls each generation step, and how its multi-hop attention mechanism helps in learning correlations between memories.
GraPPa: Grammar-Augmented Pre-Training for Table Semantic Parsing
- Tao Yu, Chien-Sheng Wu, Caiming Xiong
- Computer ScienceInternational Conference on Learning…
- 29 September 2020
GraPPa is an effective pre-training approach for table semantic parsing that learns a compositional inductive bias in the joint representations of textual and tabular data and significantly outperforms RoBERTa-large as the feature representation layers and establishes new state-of-the-art results on all of them.
Personalizing Dialogue Agents via Meta-Learning
- Zhaojiang Lin, Andrea Madotto, Chien-Sheng Wu, Pascale Fung
- Computer ScienceAnnual Meeting of the Association for…
- 1 May 2019
This paper proposes to extend Model-Agnostic Meta-Learning (MAML) to personalized dialogue learning without using any persona descriptions, and demonstrates that its model outperforms non-meta-learning baselines using automatic evaluation metrics, and in terms of human-evaluated fluency and consistency.
Discriminative Nearest Neighbor Few-Shot Intent Detection by Transferring Natural Language Inference
- Jianguo Zhang, Kazuma Hashimoto, Caiming Xiong
- Computer ScienceConference on Empirical Methods in Natural…
- 25 October 2020
This paper proposes to boost the discriminative ability by transferring a natural language inference (NLI) model, and achieves more stable and accurate in-domain and OOS detection accuracy than RoBERTa-based classifiers and embedding-based nearest neighbor approaches.
UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models
- Tianbao Xie, Chen Henry Wu, Tao Yu
- Computer ScienceConference on Empirical Methods in Natural…
- 16 January 2022
The UnifiedSKG framework is proposed, which unifies 21 SKG tasks into a text-to-text format, aiming to promote systematic SKG research, instead of being exclusive to a single task, domain, or dataset.
...
...