NSTL国家科技图书文献中心

1. SafeLLMs: A Benchmark for Secure Bilingual Evaluation of Large Language Models NSTL国家科技图书文献中心

Wenhan Liang | Huijia Wu ... - 《Natural Language Processing and Chinese Computing,Part II》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 437～448 - 共12页

摘要：Since the advent of the GPT-3.5 model | , numerous large language models (LLMs) have emerged in | China. With the increasing number of users, the | security of these models has garnered extensive attention | from researchers. However, the current evaluation

关键词： Model evaluation | Security assessment | Large language models

2. TARGET: Temp late-Transferable Backdoor Attack Against Prompt-Based NLP Models via GPT4 NSTL国家科技图书文献中心

Zihao Tan | Qingliang Chen ... - 《Natural Language Processing and Chinese Computing,Part II》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 398～411 - 共14页

摘要：Prompt-based learning has been widely applied | in many low-resource natural language processing | (NLP) tasks such as few-shot scenarios. However, this | paradigm has been shown to be vulnerable to backdoor | attacks. Most of the existing attack methods focus on

关键词： Prompt-based learning | Backdoor attack | Few-shot classification tasks

3. ConFit: Contrastive Fine-Tuning of Text-to-Text Transformer for Relation Classification NSTL国家科技图书文献中心

Jiaxin Duan | Fengyu Lu ... - 《Natural Language Processing and Chinese Computing,Part II》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 16～29 - 共14页

摘要：Relation classification (RC) is commonly the | second step in a relation extraction pipeline, which | asserts the relation of two identified entities based on | their context. The latest trend for dealing with the | task resorts to pre-trained language models (PLMs

关键词： Relation classification | Fine-tuning | Contrastive learning

4. DRAK: Unlocking Molecular Insights with Domain-Specific Retrieval-Augmented Knowledge in LLMs NSTL国家科技图书文献中心

Jinzhe Liu | Xiangsheng Huang ... - 《Natural Language Processing and Chinese Computing,Part II》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 255～267 - 共13页

摘要：Large Language Models (LLMs) typically | manifest knowledge gap in specialized applications due to | pre-training on generalized textual corpora | . Although fine-tuning and modality alignment aim to bridge | this gap, their inability to provide comprehensive

关键词： Retrieval-augmented knowledge | Knowledge injection | Biomolecular domain

5. Prompt Debiasing via Causal Intervention for Event Argument Extraction NSTL国家科技图书文献中心

Jiaju Lin | Jie Zhou ... - 《Natural Language Processing and Chinese Computing,Part II》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 96～108 - 共13页

摘要：Prompt-based methods have become increasingly | popular among information extraction tasks (e.g., event | argument extraction), especially in low-data scenarios | . By formatting a fine-tuning task into a pre | -training objective, prompt-based methods resolve the data

关键词： Event argument extraction | Prompt learning | Causal intervention

6. CouBRE: Counterfactual NLI For Low-Resource Biomedical Relation Extraction NSTL国家科技图书文献中心

Jin Zhong | Yongbin Liu ... - 《Natural Language Processing and Chinese Computing,Part II》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 109～121 - 共13页

摘要：Biomedical relation extraction is a core | problem in biomedical natural language processing, whose | goal is to classify the relations between entity | mentions within a classified given text, being modeled as | a classification method. There has been some

关键词： Biomedical relation extraction | Natural language inference | Counterfactual inference | Debias

7. HTCSI: A Hierarchical Text Classification Method Based on Selection-Inference NSTL国家科技图书文献中心

Yiming Xu | Jianzhou Feng ... - 《Natural Language Processing and Chinese Computing,Part II》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 307～318 - 共12页

摘要：Hierarchical Text Classification (HTC) is a | complex task that involves categorizing text within a | hierarchical labeling system. Current methodologies primarily | focus on integrating hierarchical information into | encoding models to achieve enhanced text representations

关键词： Hierarchical text classification | Large language models | Chain of thought

8. Bread: A Hybrid Approach for Instruction Data Mining Through Balanced Retrieval and Dynamic Data Sampling NSTL国家科技图书文献中心

Xinlin Zhuang | Xin Mao ... - 《Natural Language Processing and Chinese Computing,Part II》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 229～240 - 共12页

摘要：Recent advancements in Instruction Tuning (IT | ) have shown promise for aligning Large Language Models | (LLMs) with users' intentions, yet its efficacy is | often compromised by dependence on high-quality | datasets. Previous works have concentrated on the

关键词： Large language models | Instruction tuning | Data selection

9. Span-Based Chinese Few-Shot NER with Contrastive and Prompt Learning NSTL国家科技图书文献中心

Feiyang Ye | Peichao Lai ... - 《Natural Language Processing and Chinese Computing,Part II》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 43～55 - 共13页

摘要：For Chinese Named Entity Recognition (NER | ) tasks, achieving better performance with fewer | training samples remains a challenge. Previous works | primarily focus on enhancing model performance in NER by | incorporating additional knowledge to construct entity

关键词： Named entity recognition | Contrastive learning | Few-shot learning

10. Retrieval-Augmented Code Generation for Universal Information Extraction NSTL国家科技图书文献中心

Yucan Guo | Zixuan Li ... - 《Natural Language Processing and Chinese Computing,Part II》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 30～42 - 共13页

摘要：Information Extraction (IE) aims to extract | structural knowledge (e.g., entities, relations, events | ) from natural language texts. Recently, Large Language | Models (LLMs) with code-style prompts have demonstrated | powerful capabilities in IE tasks. However, adopting code

摘要：Since the advent of the GPT-3.5 model | , numerous large language models (LLMs) have emerged in | China. With the increasing number of users, the | security of these models has garnered extensive attention | from researchers. However, the current evaluation

关键词： Model evaluation | Security assessment | Large language models

摘要：Prompt-based learning has been widely applied | in many low-resource natural language processing | (NLP) tasks such as few-shot scenarios. However, this | paradigm has been shown to be vulnerable to backdoor | attacks. Most of the existing attack methods focus on

关键词： Prompt-based learning | Backdoor attack | Few-shot classification tasks

摘要：Relation classification (RC) is commonly the | second step in a relation extraction pipeline, which | asserts the relation of two identified entities based on | their context. The latest trend for dealing with the | task resorts to pre-trained language models (PLMs

关键词： Relation classification | Fine-tuning | Contrastive learning

摘要：Large Language Models (LLMs) typically | manifest knowledge gap in specialized applications due to | pre-training on generalized textual corpora | . Although fine-tuning and modality alignment aim to bridge | this gap, their inability to provide comprehensive

关键词： Retrieval-augmented knowledge | Knowledge injection | Biomolecular domain

摘要：Prompt-based methods have become increasingly | popular among information extraction tasks (e.g., event | argument extraction), especially in low-data scenarios | . By formatting a fine-tuning task into a pre | -training objective, prompt-based methods resolve the data

关键词： Event argument extraction | Prompt learning | Causal intervention

摘要：Biomedical relation extraction is a core | problem in biomedical natural language processing, whose | goal is to classify the relations between entity | mentions within a classified given text, being modeled as | a classification method. There has been some

关键词： Biomedical relation extraction | Natural language inference | Counterfactual inference | Debias

摘要：Hierarchical Text Classification (HTC) is a | complex task that involves categorizing text within a | hierarchical labeling system. Current methodologies primarily | focus on integrating hierarchical information into | encoding models to achieve enhanced text representations

关键词： Hierarchical text classification | Large language models | Chain of thought

摘要：Recent advancements in Instruction Tuning (IT | ) have shown promise for aligning Large Language Models | (LLMs) with users' intentions, yet its efficacy is | often compromised by dependence on high-quality | datasets. Previous works have concentrated on the

关键词： Large language models | Instruction tuning | Data selection

摘要：For Chinese Named Entity Recognition (NER | ) tasks, achieving better performance with fewer | training samples remains a challenge. Previous works | primarily focus on enhancing model performance in NER by | incorporating additional knowledge to construct entity

关键词： Named entity recognition | Contrastive learning | Few-shot learning

摘要：Information Extraction (IE) aims to extract | structural knowledge (e.g., entities, relations, events | ) from natural language texts. Recently, Large Language | Models (LLMs) with code-style prompts have demonstrated | powerful capabilities in IE tasks. However, adopting code

关键词： Information extraction | Code generation | Large language model | Information retrieval

4008-161-200 800-990-8900

国家科技图书文献中心