NSTL国家科技图书文献中心

1. Structure-aware Fine-tuning for Code Pre-trained Models NSTL国家科技图书文献中心

Jiayi Wu | Renyu Zhu ... - 《2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation,Vol. 19》 - Joint International Conference on Computational Linguistics, Language Resources and Evaluation - 2024, - 15362～15372 - 共11页

摘要：Over the past few years, we have witnessed remarkable advancements in Code Pre-trained Models (CodePTMs). These models achieved excellent representation capabilities by designing structure-based pre-t...

关键词： Pre-traind Language Models | Model Tuning | Code Structure

2. Make Prompt-based Black-Box Tuning Colorful: Boosting Model Generalization from Three Orthogonal Perspectives NSTL国家科技图书文献中心

Qiushi Sun | Chengcheng Han ... - 《2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation,Vol. 14》 - Joint International Conference on Computational Linguistics, Language Resources and Evaluation - 2024, - 10958～10969 - 共12页

摘要：Large language models (LLMs) have shown increasing power on various natural language processing (NLP) tasks. However, tuning these models for downstream tasks usually needs exorbitant costs or is unav...

关键词： Black-box Language Models | Derivative-free Optimization | Parameter-Efficient Tuning

3. Conjoin After Decompose: Improving Few-Shot Performance of Named Entity Recognition NSTL国家科技图书文献中心

Chengcheng Han | Renyu Zhu ... - 《2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation,Vol. 5》 - Joint International Conference on Computational Linguistics, Language Resources and Evaluation - 2024, - 3707～3717 - 共11页

摘要：Prompt-based methods have been widely used in few-shot named entity recognition (NER). In this paper, we first conduct a preliminary experiment and observe that the key to affecting the performance of...

关键词： named entity recognition | prompt learning | BART

4. Structure clearance design in wind tunnel tests with implications for aerodynamic drag of high-speed trains NSTL国家科技图书文献中心

Zhixiang Huang | Hanjie Huang ... - 《International Journal of Rail Transportation》 - 2023,11(6) - 961～972 - 共12页

摘要：ABSTRACT The influences of vestibule diaphragm gap, wheel-rail clearance, and strut-plate gap on the aerodynamic drag of a 1/8th-scale HST model were investigated in an 8?m?×?6?m wind tunnel. The Reyn...

关键词： High-Speed train | structure clearance | aerodynamic drag | Reynolds number effect | wind tunnel test

5. Programming Knowledge Tracing: A Comprehensive Dataset and A New Model NSTL国家科技图书文献中心

Renyu Zhu | Dongxiang Zhang ... - 《2022 IEEE International Conference on Data Mining Workshops: 22nd IEEE International Conference on Data Mining Workshops (ICDMW), 28 November-1 December 2022, Orlando, Florida》 - IEEE International Conference on Data Mining Workshops - 2022, - 298～307 - 共10页

摘要：In this paper, we study knowledge tracing in the domain of programming education and make two important contributions. First, we harvest and publish so far the most comprehensive dataset, namely BePKT...

关键词： Codes | Conferences | Education | Predictive models | Data models | Behavioral sciences | Data mining

6. On Disambiguating Authors: Collaboration Network Reconstruction in a Bottom-up Manner NSTL国家科技图书文献中心

Na Li | Renyu Zhu ... - 《2021 IEEE 37th International Conference on Data Engineering: ICDE 2021, Chania, Greece, 19-22 April 2021, [v.2]》 - IEEE International Conference on Data Engineering - 2021, - 888～899 - 共12页

摘要：Author disambiguation arises when different authors share the same name, which is a critical task in digital libraries, such as DBLP, CiteULike, CiteSeerX, etc. While the state-of-the-art methods have...

关键词： Collaboration network | Author disambiguation | Probabilistic generative model | Exponential family

7. On Disambiguating Authors: Collaboration Network Reconstruction in a Bottom-up Manner NSTL国家科技图书文献中心

Na Li | Renyu Zhu ... - 《2021 IEEE 37th International Conference on Data Engineering: IEEE 37th International Conference on Data Engineering (ICDE 2021), 19-22 April 2021, Virtual, Chania, Greece》 - International Conference on Data Engineering - 2021, - 888～899 - 共12页

摘要：Author disambiguation arises when different authors share the same name, which is a critical task in digital libraries, such as DBLP, CiteULike, CiteSeerX, etc. While the state-of-the-art methods have...

关键词： Conferences | Computational modeling | Collaboration | Probabilistic logic | Data engineering | Libraries | Data models

NSTL主题词： bottom-up | Writers | Network

8. When Gradient Descent Meets Derivative-Free Optimization: A Match Made in Black-Box Scenario NSTL国家科技图书文献中心

Chengcheng Han | Liqing Cui ... - 《Findings of the Association for Computational Linguistics: ACL 2023, vol. 2: 61st annual conference of the Association for Computational Linguistics, 9-14 July 2023, hybrid format》 - Annual conference of the Association for Computational Linguistics - 2023, - 868～880 - 共13页

摘要：Large pre-trained language models (PLMs) have garnered significant attention for their versatility and potential for solving a wide spectrum of natural language processing (NLP) tasks. However, the co...

9. A Neural Network Architecture for Program Understanding Inspired by Human Behaviors NSTL国家科技图书文献中心

Renyu Zhu | Lei Yuan ... - 《60th annual meeting of the Association for Computational Linguistics. Long papers, vol. 7: 60th annual meeting of the Association for Computational Linguistics (ACL 2022), 22-27 May 2022, Dublin, Ireland》 - Annual meeting of the Association for Computational Linguistics - 2022, - 5142～5153 - 共12页

摘要：Program understanding is a fundamental task in program language processing. Despite the success, existing works fail to take human behaviors as reference in understanding programs. In this paper, we c...

10. CAT-probing: A Metric-based Approach to Interpret How Pre-trained Models for Programming Language Attend Code Structure NSTL国家科技图书文献中心

Nuo Chen | Qiushi Sun ... - 《Findings of the Association for Computational Linguistics: EMNLP 2022, vol. 6: Conference on Empirical Methods in Natural Language Processing (EMNLP 2022), 7-11 November 2022, Abu Dhabi, United Arab Emirates》 - Conference on Empirical Methods in Natural Language Processing - 2022, - 4029～4037 - 共9页

摘要：Code pre-trained models (CodePTMs) have recently demonstrated significant success in code intelligence. To interpret these models, some probing methods have been applied. However, these methods fail t...

摘要：Over the past few years, we have witnessed remarkable advancements in Code Pre-trained Models (CodePTMs). These models achieved excellent representation capabilities by designing structure-based pre-t...

关键词： Pre-traind Language Models | Model Tuning | Code Structure

摘要：Large language models (LLMs) have shown increasing power on various natural language processing (NLP) tasks. However, tuning these models for downstream tasks usually needs exorbitant costs or is unav...

关键词： Black-box Language Models | Derivative-free Optimization | Parameter-Efficient Tuning

摘要：Prompt-based methods have been widely used in few-shot named entity recognition (NER). In this paper, we first conduct a preliminary experiment and observe that the key to affecting the performance of...

关键词： named entity recognition | prompt learning | BART

摘要：ABSTRACT The influences of vestibule diaphragm gap, wheel-rail clearance, and strut-plate gap on the aerodynamic drag of a 1/8th-scale HST model were investigated in an 8?m?×?6?m wind tunnel. The Reyn...

关键词： High-Speed train | structure clearance | aerodynamic drag | Reynolds number effect | wind tunnel test

摘要：In this paper, we study knowledge tracing in the domain of programming education and make two important contributions. First, we harvest and publish so far the most comprehensive dataset, namely BePKT...

关键词： Codes | Conferences | Education | Predictive models | Data models | Behavioral sciences | Data mining

摘要：Author disambiguation arises when different authors share the same name, which is a critical task in digital libraries, such as DBLP, CiteULike, CiteSeerX, etc. While the state-of-the-art methods have...

关键词： Collaboration network | Author disambiguation | Probabilistic generative model | Exponential family

摘要：Author disambiguation arises when different authors share the same name, which is a critical task in digital libraries, such as DBLP, CiteULike, CiteSeerX, etc. While the state-of-the-art methods have...

关键词： Conferences | Computational modeling | Collaboration | Probabilistic logic | Data engineering | Libraries | Data models

NSTL主题词： bottom-up | Writers | Network

摘要：Large pre-trained language models (PLMs) have garnered significant attention for their versatility and potential for solving a wide spectrum of natural language processing (NLP) tasks. However, the co...

摘要：Program understanding is a fundamental task in program language processing. Despite the success, existing works fail to take human behaviors as reference in understanding programs. In this paper, we c...

摘要：Code pre-trained models (CodePTMs) have recently demonstrated significant success in code intelligence. To interpret these models, some probing methods have been applied. However, these methods fail t...

4008-161-200 800-990-8900

国家科技图书文献中心