Adaptive Beam Search Decoding for Discrete Keyphrase Generation

Huang, Xiaoli; Xu, Tongge; Jiao, Lvan; Zu, Yueran; Zhang, Youmin

doi:10.1609/aaai.v35i14.17546

Cited by 3 publications

(3 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the molecules produced by these models lack sufficient diversity and perform poorly in generating large molecules. Huang et al proposed a molecular diffusion model (MDM) to solve these two problems . MDM enhances interatomic connections that are absent in the 3D point cloud representation of molecules and presents separated encoders to capture interatomic forces of different strengths.…”

Section: Molecular Generative Model Architecturesmentioning

confidence: 99%

“…Flow-based models are able to achieve very bright results in the task of reconstructing the input molecule. ,− (5). Diffusion-based models are well capable of generating 3D structures of molecules, achieving competitive results in tasks of generating high-precision conformations and molecule generation for target protein. ,, …”

Section: Performance Comparison Of Molecular Generative Modelsmentioning

confidence: 99%

See 1 more Smart Citation

Deep Generative Models in De Novo Drug Molecule Generation

Pang,

Qiao,

Zeng

et al. 2023

J. Chem. Inf. Model.

View full text Add to dashboard Cite

The discovery of new drugs has important implications for human health. Traditional methods for drug discovery rely on experiments to optimize the structure of lead molecules, which are timeconsuming and high-cost. Recently, artificial intelligence has exhibited promising and efficient performance for drug-like molecule generation. In particular, deep generative models achieve great success in de novo generation of drug-like molecules with desired properties, showing massive potential for novel drug discovery. In this study, we review the recent progress of molecule generation using deep generative models, mainly focusing on molecule representations, public databases, data processing tools, and advanced artificial intelligence based molecule generation frameworks. In particular, we present a comprehensive comparison of state-of-the-art deep generative models for molecule generation and a summary of commonly used molecular design strategies. We identify research gaps and challenges of molecule generation such as the need for better databases, missing 3D information in molecular representation, and the lack of high-precision evaluation metrics. We suggest future directions for molecular generation and drug discovery.

show abstract

Section: Molecular Generative Model Architecturesmentioning

confidence: 99%

Section: Performance Comparison Of Molecular Generative Modelsmentioning

confidence: 99%

Deep Generative Models in De Novo Drug Molecule Generation

Pang,

Qiao,

Zeng

et al. 2023

J. Chem. Inf. Model.

View full text Add to dashboard Cite

show abstract

“…Ammar et al (2017) also used semi-supervised learning for entity and relation extraction from scientific papers. Other works used supervised or semi-supervised keyphrase extraction or generation (Kulkarni et al, 2022;Caragea, 2021, 2019;Alzaidy et al, 2019;Florescu and Caragea, 2017;Ye and Wang, 2018;Garg et al, 2023Garg et al, , 2022Ye et al, 2021;Huang et al, 2021;Wu et al, 2023). Moreover, Ammar et al (2018) In contrast, we propose a cosine embedding constraint added to multi-task knowledge distillation that enforces teacher-student similarity in the embedding space.…”

Section: Related Workmentioning

confidence: 99%

Multi-Task Knowledge Distillation with Embedding Constraints for Scholarly Keyphrase Boundary Classification

Park,

Caragea

2023

Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

The task of scholarly keyphrase boundary classification aims at identifying keyphrases from scientific papers and classifying them with their types from a set of predefined classes (e.g., task, process, or material). Despite the importance of keyphrases and their types in many downstream applications including indexing, searching, and question answering over scientific documents, scholarly keyphrase boundary classification is still an under-explored task. In this work, we propose a novel embedding constraint on multitask knowledge distillation which enforces the teachers (single-task models) and the student (multi-task model) similarity in the embedding space. Specifically, we enforce that the student model is trained not only to imitate the teachers' output distribution over classes, but also to produce language representations that are similar to those produced by the teachers. Our results show that the proposed approach outperforms previous works and strong baselines on three datasets of scientific documents.

show abstract

Seq2Set2Seq: A Two-stage Disentangled Method for Reply Keyword Generation in Social Media

Liu,

Li,

et al. 2024

ACM Trans. Asian Low-Resour. Lang. Inf. Process.

View full text Add to dashboard Cite

Social media produces large amounts of contents every day. How to predict the potential influences of the contents from a social reply feedback perspective is a key issue that has not been explored. Thus, we propose a novel task named reply keyword prediction in social media, which aims to predict the keywords in the potential replies as many aspects as possible. One prerequisite challenge is that the accessible social media datasets labeling such keywords remain absent. To solve this issue, we propose a new dataset, to study the reply keyword prediction in Social Media. This task could be seen as a single-turn dialogue keyword prediction for open-domain dialogue system. However, existing methods for dialogue keyword prediction cannot be adopted directly, which have two main drawbacks. First, they do not provide an explicit mechanism to model topic complementarity between keywords which is crucial in social media to controllably model all aspects of replies. Second, the collocations of keywords are not explicitly modeled, which also makes it less controllable to optimize for fine-grained prediction since the context information is much less than that in dialogue. To address these issues, we propose a two-stage disentangled framework, which can optimize the complementarity and collocation explicitly in a disentangled fashion. In the first stage, we use a sequence-to-set paradigm via multi-label prediction and determinantal point processes, to generate a set of keyword seeds satisfying the complementarity. In the second stage, we adopt a set-to-sequence paradigm via seq2seq model with the keyword seeds guidance from the set, to generate the more-fine-grained keywords with collocation. Experiments show that this method can generate not only a more diverse set of keywords but also more relevant and consistent keywords. Furthermore, the keywords obtained based on this method can achieve better reply generation results in the retrieval-based system than others.

show abstract

Adaptive Beam Search Decoding for Discrete Keyphrase Generation

Cited by 3 publications

References 23 publications

Deep Generative Models in De Novo Drug Molecule Generation

Deep Generative Models in De Novo Drug Molecule Generation

Multi-Task Knowledge Distillation with Embedding Constraints for Scholarly Keyphrase Boundary Classification

Seq2Set2Seq: A Two-stage Disentangled Method for Reply Keyword Generation in Social Media

Contact Info

Product

Resources

About