DualTKB: A Dual Learning Bridge between Text and Knowledge Base

Dognin, Pierre; Melnyk, Igor; Padhi, Inkit; Santos, Cícero Nogueira dos; Das, Payel

doi:10.18653/v1/2020.emnlp-main.694

Cited by 11 publications

(9 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Lin et al (2020); Liu et al (2021) study CommonGen, which aims to generate coherent sentences containing the given common concepts. Dognin et al (2020); Agarwal et al (2021) study the data-to-text generation (Kukich, 1983), which aims to convert facts in KGs into natural language. None of these works meet the requirements on openness and interpretability.…”

Section: Related Workmentioning

confidence: 99%

DKG: A Descriptive Knowledge Graph for Explaining Relationships between Entities

Huang¹,

Zhu²,

Chang³

et al. 2022

Preprint

View full text Add to dashboard Cite

We propose Descriptive Knowledge Graph (DKG) -an open and interpretable form of modeling relationships between entities. In DKGs, relationships between entities are represented by relation descriptions. For instance, the relationship between entities of machine learning and algorithm can be described as "Machine learning explores the study and construction of algorithms that can learn from and make predictions on data." To construct DKGs, we propose a self-supervised learning method to extract relation descriptions with the analysis of dependency patterns and a transformer-based relation description synthesizing model to generate relation descriptions. Experiments demonstrate that our system can extract and generate high-quality relation descriptions for explaining entity relationships.

show abstract

Section: Related Workmentioning

confidence: 99%

DKG: A Descriptive Knowledge Graph for Explaining Relationships between Entities

Huang¹,

Zhu²,

Chang³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…For the samples in D tr , which have no entity labels, we do not calculate their loss during entity generation, but instead calculate their loss in response generation ( §3.4) to realize end-to-end optimization like DualTKB (Dognin et al, 2020).…”

Section: Autoregressive Entity Generationmentioning

confidence: 99%

Autoregressive Entity Generation for End-to-End Task-Oriented Dialog

Huang¹,

Wang²

2022

Preprint

View full text Add to dashboard Cite

Task-oriented dialog (TOD) systems often require interaction with an external knowledge base to retrieve necessary entity (e.g., restaurant) information to support the response generation. Most current end-to-end TOD systems either retrieve the KB information explicitly or embed it into model parameters for implicit access. While the former approach demands scanning the KB at each turn of response generation, which is inefficient when the KB scales up, the latter approach shows higher flexibility and efficiency. In either approach, the systems may generate a response with conflicting entity information. To address this issue, we propose to generate the entity autoregressively first and leverage it to guide the response generation in an end-to-end system. To ensure entity consistency, we impose a trie constraint on entity generation. We also introduce a logit concatenation strategy to facilitate gradient backpropagation for end-to-end training. Experiments on MultiWOZ 2.1 single and CAMREST show that our system can generate more high-quality and entity-consistent responses.

show abstract

“…Recently, Dognin et al (2020); Guo et al (2020bGuo et al ( , 2021 proposed models trained to generate in both T2G and G2T directions, with consistency cycles created to enable the use of unsupervised datasets.…”

Section: Related Workmentioning

confidence: 99%

ReGen: Reinforcement Learning for Text and Knowledge Base Generation using Pretrained Language Models

Dognin¹,

Padhi²,

Melnyk³

et al. 2021

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

Self Cite

View full text Add to dashboard Cite

Automatic construction of relevant Knowledge Bases (KBs) from text, and generation of semantically meaningful text from KBs are both long-standing goals in Machine Learning. In this paper, we present ReGen, a bidirectional generation of text and graph leveraging Reinforcement Learning (RL) to improve performance. Graph linearization enables us to re-frame both tasks as a sequence to sequence generation problem regardless of the generative direction, which in turn allows the use of Reinforcement Learning for sequence training where the model itself is employed as its own critic leading to Self-Critical Sequence Training (SCST). We present an extensive investigation demonstrating that the use of RL via SCST benefits graph and text generation on WebNLG+ 2020 and TEKGEN datasets.Our system provides state-of-the-art results on WebNLG+ 2020 by significantly improving upon published results from the WebNLG 2020+ Challenge for both text-to-graph and graph-to-text generation tasks. More details in https://github.com/IBM/regen.

show abstract

DualTKB: A Dual Learning Bridge between Text and Knowledge Base

Cited by 11 publications

References 13 publications

DKG: A Descriptive Knowledge Graph for Explaining Relationships between Entities

DKG: A Descriptive Knowledge Graph for Explaining Relationships between Entities

Autoregressive Entity Generation for End-to-End Task-Oriented Dialog

ReGen: Reinforcement Learning for Text and Knowledge Base Generation using Pretrained Language Models

Contact Info

Product

Resources

About