Document-level event argument extraction (DEAE) aims to structure unstructured documents by identifying event participants, supporting various downstream tasks. Traditional methods rely on fine-tuning pre-trained models, demanding substantial data and complex architectures. In-context learning (ICL) provides a promising solution. However, its application in DEAE tasks faces two challenges: (1) inadequate strategies for selecting examples and (2) insufficient integration of error feedback from large language models (LLMs). To address these challenges, we introduce GEIES (Guidelines-Enhanced ICL with Examples Selection), a novel ICL framework. It uses a dual similarity selector (DSS) to select diverse and relevant examples. Additionally, GEIES employs a guidelines generator (GG) to convert LLM error feedback into actionable guidelines. Finally, we embed the examples and guidelines into the prompt context. Experiments confirm GEIES’s effectiveness in DEAE, with ablation studies validating the contributions of its components.

In the field of ancient Chinese text, extracting and analysing temporal and geographic information are crucial for understanding the personal experiences of historical figures, the development of historical events, and the overall historical background. Currently, named entity recognition(NER) strategies such as BERT+CRF are used to extract temporal and geographic information from ancient Chinese text. However, ancient Chinese text covers a vast time span, and the temporal and geographic entities constantly evolve and change, making it difficult to extract these entities from text. This paper proposes a temporal and geographic extraction model for ancient Chinese text, enhanced by time-series external knowledge base. The extraction of proprietary nouns and general structures are divided into two independent networks. An external database is applied to enhance extraction of proprietary nouns and reduce noise for general structure inference. We constructed address trees and chronological tables containing commonly used places and time-related keywords from different periods and collected 12,000 texts spanning 3,000 years for extensive training. Overall, our research highlights the importance of external knowledge base for ancient Chinese NER, and provides new ideas for research in related fields.