NSTL国家科技图书文献中心

1. Diverse Text-to-3D Synthesis with Augmented Text Embedding NSTL国家科技图书文献中心

Uy Dieu Tran | Minh Luu ... - 《Computer Vision - ECCV 2024,Part LXXV》 - European Conference on Computer Vision - 2025, - 217～235 - 共19页

摘要：Text-to-3D synthesis has recently emerged as a | pretrained text-to-image models as guiding visual priors | existing text-to-3D methods is that 3D models obtained | the same text* prompt. We propose to use augmented* | text prompts via textual inversion of reference

关键词： Text-to-3D | 3D computer vision | Generative models

2. AttRel: Single Module Based Joint Entity and Relation Extraction with Attention Enhanced Text Embedding NSTL国家科技图书文献中心

Mengmeng Cui | Chenbin Li ... - 《Advanced Data Mining and Applications,Part V》 - International Conference on Advanced Data Mining and Applications - 2025, - 328～343 - 共16页

摘要： text embedding vectors to improve the accuracy of | joint extraction approach with attention enhanced text | embedding, named AttRel. Specifically, a novel single | Information extraction is a vital subtask of | knowledge graph constructions, where the joint extraction

关键词： Joint extraction | Attention mechanisms | Relation embedding

3. Interpretable prediction of drug-drug interactions via text embedding in biomedical literature NSTL国家科技图书文献中心

Jung S. | Yoo S. - 《Computers in Biology and Medicine》 - 2025,185 - Article 109496～Article 109496 - 共11页

摘要： two components: drug embedding* and DDI prediction* | . The drug embedding* module extracts representation* | sentence and sequence embedding* methods. For sentence* | embedding, a pre-trained biomedical language model is used | . For sequence embedding, sentence embedding* vectors*

关键词： Attention mechanism | Deep learning | Drug embedding | Drug expression | Drug-drug interaction

4. Regulating the level of manipulation in text augmentation with systematic adjustment and advanced sentence embedding NSTL国家科技图书文献中心

Yuho,Cha | Younghoon,Lee - 《Neural computing & applications》 - 2025,37(5) - 3473～3487 - 共15页

摘要：Text augmentation, a method for generating | the most important issue in text* augmentation; low* | proposes a systematically adjustable text* augmentation* | sentence embedding* methodology to achieve robust pseudo* | leverage combined sentence embedding, which incorporates

关键词： Text augmentation | The level of manipulation | Advanced sentence embedding | Reliable pseudo-labels

5. PlanBERT: From Messy Zonal Plans to Informative Vector Embeddings NSTL国家科技图书文献中心

Henrik Bradland | Morten Goodwin ... - 《Artificial Intelligence XLI,Part I》 - SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence - 2025, - 175～188 - 共14页

摘要：Text embedding models trained on vast web | fine-tuning of embedding* models. The approach builds* | (LLMs), and decorrelation of embedding* features. The* | -scraped corpus generalize well to daily language | . However, they often fall short when applied in

关键词： Text embedding | Feature decorrelation | Domain-adaptation

6. Integrating Embedding and LSHiForest in English Text Anomaly Detection NSTL国家科技图书文献中心

Qingquan Tong | Rongju Yao - 《Concurrency and computation: practice and experience》 - 2025,37(3) - e8370.1～e8370.11 - 共11页

摘要： and sparse nature of text* data, leading to* | ) into the process of English text* anomaly detection* | transforming English text* into feature vectors, followed by* | across various text* datasets. The effectiveness of this* | English text, particularly in scenarios involving large

关键词： anomaly detection | English Dext | LSHiForest | outlier identification | word embedding

7. Visually guided extraction of prevalent topics NSTL国家科技图书文献中心

Daniel Witschard | Ilir Jusufi ... - 《Information visualization》 - 2025,24(2) - 179～198 - 共20页

摘要：The sensemaking process of large sets of text | large text* corpora exist, many of them lack the* | two text* corpora with very different content. We* | that our approach is suitable for text* mining, that* | documents is highly challenging for tasks such as

关键词： Visual analytics | text mining | text embedding | topic modeling | similarity calculations

8. A Transformer-Based Tabular Approach to Detect Toxic Comments NSTL国家科技图书文献中心

Ghivvago Damas | Rafael Torres Anchie...... - 《Intelligent Systems,Part IV》 - Brazilian Conference on Intelligent Systems - 2025, - 18～30 - 共13页

摘要： text-embedding-3-large model. | using modern embedding models as language embedders | In recent years, there has been a significant | increase in toxic and hateful speech on social media | platforms, becoming deeply entrenched in online

关键词： Toxic and hateful speech | Deep learning | FT-Transformer | Embedding models | Text classification

9. NLWM: A Robust, Efficient and High-Quality Watermark for Large Language Models NSTL国家科技图书文献中心

Mengting Song | Ziyuan Li ... - 《Web Information Systems Engineering - WISE 2024,Part V》 - International Conference on Web Information Systems Engineering - 2025, - 320～335 - 共16页

摘要： challenge. Embedding* watermarks during text generation is* | model-generated and human-created text* a significant* | text quality and robustness, often struggling to | achieve high accuracy over limited text* sequences. Hence* | in text* quality observed in current methods, we*

关键词： Large language models | Text watermarking | N-bit watermark | Watermark embedding and extraction | BCH code

10. Retrieval-Augmented Generation Architecture Framework: Harnessing the Power of RAG NSTL国家科技图书文献中心

Richard Shan | Tony Shan - 《Cognitive Computing - ICCC 2024》 - International Conference on Cognitive Computing | Services Conference Federation - 2025, - 88～104 - 共17页

摘要： databases, knowledge bases, text* processing, frontend* | This paper presents a comprehensive | exploration of the Retrieval-Augmented Generation | Architecture Framework (RAGAF), structured around seven key | modules: Generator, Retriever, Orchestration, UI, Source

关键词： Retrieval-Augmented generation | RAG architecture | Framework | Natural language processing | Generative models | Large language models | Text embedding | Vector database | Multimodal data integration | Evaluation metrics

摘要：Text-to-3D synthesis has recently emerged as a | pretrained text-to-image models as guiding visual priors | existing text-to-3D methods is that 3D models obtained | the same text prompt. We propose to use augmented | text prompts via textual inversion of reference

关键词： Text-to-3D | 3D computer vision | Generative models

摘要： text embedding vectors to improve the accuracy of | joint extraction approach with attention enhanced text | embedding, named AttRel. Specifically, a novel single | Information extraction is a vital subtask of | knowledge graph constructions, where the joint extraction

关键词： Joint extraction | Attention mechanisms | Relation embedding

摘要： two components: drug embedding and DDI prediction | . The drug embedding module extracts representation | sentence and sequence embedding methods. For sentence | embedding, a pre-trained biomedical language model is used | . For sequence embedding, sentence embedding vectors

关键词： Attention mechanism | Deep learning | Drug embedding | Drug expression | Drug-drug interaction

摘要：Text augmentation, a method for generating | the most important issue in text augmentation; low | proposes a systematically adjustable text augmentation | sentence embedding methodology to achieve robust pseudo | leverage combined sentence embedding, which incorporates

关键词： Text augmentation | The level of manipulation | Advanced sentence embedding | Reliable pseudo-labels

摘要：Text embedding models trained on vast web | fine-tuning of embedding models. The approach builds | (LLMs), and decorrelation of embedding features. The | -scraped corpus generalize well to daily language | . However, they often fall short when applied in

关键词： Text embedding | Feature decorrelation | Domain-adaptation

摘要： and sparse nature of text data, leading to | ) into the process of English text anomaly detection | transforming English text into feature vectors, followed by | across various text datasets. The effectiveness of this | English text, particularly in scenarios involving large

关键词： anomaly detection | English Dext | LSHiForest | outlier identification | word embedding

摘要：The sensemaking process of large sets of text | large text corpora exist, many of them lack the | two text corpora with very different content. We | that our approach is suitable for text mining, that | documents is highly challenging for tasks such as

关键词： Visual analytics | text mining | text embedding | topic modeling | similarity calculations

摘要： text-embedding-3-large model. | using modern embedding models as language embedders | In recent years, there has been a significant | increase in toxic and hateful speech on social media | platforms, becoming deeply entrenched in online

关键词： Toxic and hateful speech | Deep learning | FT-Transformer | Embedding models | Text classification

摘要： challenge. Embedding watermarks during text generation is | model-generated and human-created text a significant | text quality and robustness, often struggling to | achieve high accuracy over limited text sequences. Hence | in text quality observed in current methods, we

关键词： Large language models | Text watermarking | N-bit watermark | Watermark embedding and extraction | BCH code

摘要： databases, knowledge bases, text processing, frontend | This paper presents a comprehensive | exploration of the Retrieval-Augmented Generation | Architecture Framework (RAGAF), structured around seven key | modules: Generator, Retriever, Orchestration, UI, Source

关键词： Retrieval-Augmented generation | RAG architecture | Framework | Natural language processing | Generative models | Large language models | Text embedding | Vector database | Multimodal data integration | Evaluation metrics

4008-161-200 800-990-8900

国家科技图书文献中心

摘要：Text-to-3D synthesis has recently emerged as a | pretrained text-to-image models as guiding visual priors | existing text-to-3D methods is that 3D models obtained | the same text* prompt. We propose to use augmented* | text prompts via textual inversion of reference

摘要： two components: drug embedding* and DDI prediction* | . The drug embedding* module extracts representation* | sentence and sequence embedding* methods. For sentence* | embedding, a pre-trained biomedical language model is used | . For sequence embedding, sentence embedding* vectors*

摘要：Text augmentation, a method for generating | the most important issue in text* augmentation; low* | proposes a systematically adjustable text* augmentation* | sentence embedding* methodology to achieve robust pseudo* | leverage combined sentence embedding, which incorporates

摘要：Text embedding models trained on vast web | fine-tuning of embedding* models. The approach builds* | (LLMs), and decorrelation of embedding* features. The* | -scraped corpus generalize well to daily language | . However, they often fall short when applied in

摘要： and sparse nature of text* data, leading to* | ) into the process of English text* anomaly detection* | transforming English text* into feature vectors, followed by* | across various text* datasets. The effectiveness of this* | English text, particularly in scenarios involving large

摘要：The sensemaking process of large sets of text | large text* corpora exist, many of them lack the* | two text* corpora with very different content. We* | that our approach is suitable for text* mining, that* | documents is highly challenging for tasks such as

摘要： challenge. Embedding* watermarks during text generation is* | model-generated and human-created text* a significant* | text quality and robustness, often struggling to | achieve high accuracy over limited text* sequences. Hence* | in text* quality observed in current methods, we*

摘要： databases, knowledge bases, text* processing, frontend* | This paper presents a comprehensive | exploration of the Retrieval-Augmented Generation | Architecture Framework (RAGAF), structured around seven key | modules: Generator, Retriever, Orchestration, UI, Source