NSTL国家科技图书文献中心

1. Generalization potential of large language models NSTL国家科技图书文献中心

Mikhail,Budnikov | Anna,Bykova ... - 《Neural computing & applications》 - 2025,37(4) - 1973～1997 - 共25页

摘要： especially the advent of large* language models (LLMs* | language models with stochastic parrots to stating that | language processing tasks. | The rise of deep learning techniques and | ) intensified the discussions around possibilities that

关键词： Large language models | Generalization | Semantic information

2. Column Property Annotation Using Large Language Models NSTL国家科技图书文献中心

Keti Korini | Christian Bizer - 《The Semantic Web: ESWC 2024 Satellite Events,Part I》 - European Semantic Web Conference - 2025, - 61～70 - 共10页

摘要： large language models (LLMs) for the CPA task. We | Column property annotation (CPA), also known | as column relationship prediction, is the task of | predicting the semantic relationship between two columns in | a table given a set of candidate relationships

关键词： Table annotation | Large language models | Column property annotation

3. Unlocking the Potential of Large Language Models for Explainable Recommendations NSTL国家科技图书文献中心

Yucong Luo | Mingyue Cheng ... - 《Database Systems for Advanced Applications,Part V》 - International Conference on Database Systems for Advanced Applications | International Workshop on Big Data Management and Service | International Workshop on Graph Data Management and Analysis | International Workshop on Big Data Quality Management | Workshop on Emerging Results inData Science and Engineering - 2025, - 286～303 - 共18页

摘要： generator with the recently emerging large language models | systems focus on using small-size language* models. It* | increasingly prevalent, largely due to advances in language | recommender models* and LLM-based explanation generators* | experiments over several benchmark recommender models* and*

关键词： Explainable recommendation | Large language models

4. Enhancing Translation Validation of Compiler Transformations with Large Language Models NSTL国家科技图书文献中心

Yanzhao Wang | Fei Xie - 《International journal of software engineering and knowledge engineering》 - 2025,35(1) - 45～57 - 共13页

摘要： integrates Large* Language Models (LLMs) into translation* | This paper presents a framework that | validation, targeting LLVM compiler transformations where | formal verification tools fall short. Our framework | utilizes the existing tools, like Alive2, to perform

关键词： Large Language Models | formal verification | compilers | LLVM

5. LLMs4OM: Matching Ontologies with Large Language Models NSTL国家科技图书文献中心

Hamed Babaei Giglou | Jennifer D'Souza ... - 《The Semantic Web: ESWC 2024 Satellite Events,Part I》 - European Semantic Web Conference - 2025, - 25～35 - 共11页

摘要： of the potential of Large* Language Models (LLMs* | knowledge or predictive models, with limited exploration | Ontology Matching (OM), is a critical task in | knowledge integration, where aligning heterogeneous | ontologies facilitates data interoperability and knowledge

关键词： Ontology matching | Ontology alignment | Large language models | Retrieval augmented generation | Zero-Shot testing

6. Detoxifying Large Language Models via Kahneman-Tversky Optimization NSTL国家科技图书文献中心

Qingquan Li | Wenlong Du ... - 《Natural Language Processing and Chinese Computing,Part V》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 409～417 - 共9页

摘要：Currently, the application of Large Language | Models (LLMs) faces significant security threats | answering and language* understanding. Our proposed method* | . Harmful questions and adversarial attack prompts can | induce the LLMs to generate toxic responses. Therefore

关键词： Large language models | Detoxification | Alignment

7. Security and Privacy Challenges of Large Language Models: A Survey NSTL国家科技图书文献中心

BADHAN CHANDRA DAS | M. HADI AMINI ... - 《ACM computing surveys》 - 2025,57(6) - 152.1～152.39 - 共39页 - 被引量：2

摘要：Large language models (LLMs) have demonstrated | , such as generating and summarizing text, language | quite popular tools in natural language* processing* | advantages, these models* are also vulnerable to security* | extraordinary capabilities and contributed to multiple fields

关键词： Large language models | attack and defense mechanisms

8. Distributed Dataset Framework for Large Language Models Pre-training NSTL国家科技图书文献中心

Nao Souma | Yui Obara ... - 《Knowledge Management and Acquisition for Intelligent Systems》 - Pacific Rim Knowledge Acquisition Workshop | Pacific Rim International Conference on Artificial Intelligence - 2025, - 289～297 - 共9页

摘要：The performance of Large* Language Models (LLMs* | . Maintaining such large* datasets is challenging for a single* | ) relies on massive and high-quality pre-processed | datasets, often exceeding several terabytes in size | organization. A collaborative framework between multiple

关键词： Large language models | Distributed dataset | LLM development

9. ZeFaV: Boosting Large Language Models for Zero-Shot Fact Verification NSTL国家科技图书文献中心

Son T. Luu | Hiep Nguyen ... - 《PRICAI 2024,Part II》 - Pacific Rim International Conference on Artificial Intelligence - 2025, - 288～295 - 共8页

摘要： ability of large* language models to extract the* | language models by leveraging the in-context learning | the performance on fact verification task of large | In this paper, we propose ZeFaV - a zero-shot | based fact-checking verification framework to enhance

关键词： Fact verification | Prompting | Zero-shot | Large language models

10. On Leveraging Large Language Models for Multilingual Intent Discovery NSTL国家科技图书文献中心

RUDOLF CHOW | KING YIU SUEN ... - 《ACM transactions on management information systems》 - 2025,16(1) - 7.1～7.17 - 共17页

摘要： language models. By performing joint extraction of intent | users nat- urally change over time, models only | leveraging the multilingual capabilities of recent large | Intent discovery is vital for any real-world | dialogue systems such as chatbot. Since the intents of

摘要： especially the advent of large* language models (LLMs* | language models with stochastic parrots to stating that | language processing tasks. | The rise of deep learning techniques and | ) intensified the discussions around possibilities that

关键词： Large language models | Generalization | Semantic information

摘要： large language models (LLMs) for the CPA task. We | Column property annotation (CPA), also known | as column relationship prediction, is the task of | predicting the semantic relationship between two columns in | a table given a set of candidate relationships

关键词： Table annotation | Large language models | Column property annotation

关键词： Explainable recommendation | Large language models

摘要： integrates Large* Language Models (LLMs) into translation* | This paper presents a framework that | validation, targeting LLVM compiler transformations where | formal verification tools fall short. Our framework | utilizes the existing tools, like Alive2, to perform

关键词： Large Language Models | formal verification | compilers | LLVM

摘要： of the potential of Large* Language Models (LLMs* | knowledge or predictive models, with limited exploration | Ontology Matching (OM), is a critical task in | knowledge integration, where aligning heterogeneous | ontologies facilitates data interoperability and knowledge

关键词： Ontology matching | Ontology alignment | Large language models | Retrieval augmented generation | Zero-Shot testing

摘要：Currently, the application of Large Language | Models (LLMs) faces significant security threats | answering and language* understanding. Our proposed method* | . Harmful questions and adversarial attack prompts can | induce the LLMs to generate toxic responses. Therefore

关键词： Large language models | Detoxification | Alignment

摘要：Large language models (LLMs) have demonstrated | , such as generating and summarizing text, language | quite popular tools in natural language* processing* | advantages, these models* are also vulnerable to security* | extraordinary capabilities and contributed to multiple fields

关键词： Large language models | attack and defense mechanisms

摘要：The performance of Large* Language Models (LLMs* | . Maintaining such large* datasets is challenging for a single* | ) relies on massive and high-quality pre-processed | datasets, often exceeding several terabytes in size | organization. A collaborative framework between multiple

关键词： Large language models | Distributed dataset | LLM development

摘要： ability of large* language models to extract the* | language models by leveraging the in-context learning | the performance on fact verification task of large | In this paper, we propose ZeFaV - a zero-shot | based fact-checking verification framework to enhance

关键词： Fact verification | Prompting | Zero-shot | Large language models

摘要： language models. By performing joint extraction of intent | users nat- urally change over time, models only | leveraging the multilingual capabilities of recent large | Intent discovery is vital for any real-world | dialogue systems such as chatbot. Since the intents of

关键词： Intent discovery | multilingual | large language models

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： especially the advent of large language models (LLMs | language models with stochastic parrots to stating that | language processing tasks. | The rise of deep learning techniques and | ) intensified the discussions around possibilities that

关键词： Large language models | Generalization | Semantic information

摘要： large language models (LLMs) for the CPA task. We | Column property annotation (CPA), also known | as column relationship prediction, is the task of | predicting the semantic relationship between two columns in | a table given a set of candidate relationships

关键词： Table annotation | Large language models | Column property annotation

关键词： Explainable recommendation | Large language models

摘要： integrates Large Language Models (LLMs) into translation | This paper presents a framework that | validation, targeting LLVM compiler transformations where | formal verification tools fall short. Our framework | utilizes the existing tools, like Alive2, to perform

关键词： Large Language Models | formal verification | compilers | LLVM

摘要： of the potential of Large Language Models (LLMs | knowledge or predictive models, with limited exploration | Ontology Matching (OM), is a critical task in | knowledge integration, where aligning heterogeneous | ontologies facilitates data interoperability and knowledge

关键词： Ontology matching | Ontology alignment | Large language models | Retrieval augmented generation | Zero-Shot testing

摘要：Currently, the application of Large Language | Models (LLMs) faces significant security threats | answering and language understanding. Our proposed method | . Harmful questions and adversarial attack prompts can | induce the LLMs to generate toxic responses. Therefore

关键词： Large language models | Detoxification | Alignment

摘要：Large language models (LLMs) have demonstrated | , such as generating and summarizing text, language | quite popular tools in natural language processing | advantages, these models are also vulnerable to security | extraordinary capabilities and contributed to multiple fields

关键词： Large language models | attack and defense mechanisms

摘要：The performance of Large Language Models (LLMs | . Maintaining such large datasets is challenging for a single | ) relies on massive and high-quality pre-processed | datasets, often exceeding several terabytes in size | organization. A collaborative framework between multiple

关键词： Large language models | Distributed dataset | LLM development

摘要： ability of large language models to extract the | language models by leveraging the in-context learning | the performance on fact verification task of large | In this paper, we propose ZeFaV - a zero-shot | based fact-checking verification framework to enhance

关键词： Fact verification | Prompting | Zero-shot | Large language models

摘要： language models. By performing joint extraction of intent | users nat- urally change over time, models only | leveraging the multilingual capabilities of recent large | Intent discovery is vital for any real-world | dialogue systems such as chatbot. Since the intents of

关键词： Intent discovery | multilingual | large language models

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： especially the advent of large* language models (LLMs* | language models with stochastic parrots to stating that | language processing tasks. | The rise of deep learning techniques and | ) intensified the discussions around possibilities that

摘要： integrates Large* Language Models (LLMs) into translation* | This paper presents a framework that | validation, targeting LLVM compiler transformations where | formal verification tools fall short. Our framework | utilizes the existing tools, like Alive2, to perform

摘要： of the potential of Large* Language Models (LLMs* | knowledge or predictive models, with limited exploration | Ontology Matching (OM), is a critical task in | knowledge integration, where aligning heterogeneous | ontologies facilitates data interoperability and knowledge

摘要：Currently, the application of Large Language | Models (LLMs) faces significant security threats | answering and language* understanding. Our proposed method* | . Harmful questions and adversarial attack prompts can | induce the LLMs to generate toxic responses. Therefore

摘要：Large language models (LLMs) have demonstrated | , such as generating and summarizing text, language | quite popular tools in natural language* processing* | advantages, these models* are also vulnerable to security* | extraordinary capabilities and contributed to multiple fields

摘要：The performance of Large* Language Models (LLMs* | . Maintaining such large* datasets is challenging for a single* | ) relies on massive and high-quality pre-processed | datasets, often exceeding several terabytes in size | organization. A collaborative framework between multiple

摘要： ability of large* language models to extract the* | language models by leveraging the in-context learning | the performance on fact verification task of large | In this paper, we propose ZeFaV - a zero-shot | based fact-checking verification framework to enhance