Classical Chinese historical documents are invaluable for Chinese cultural heritage and history research, while they remain underexplored within natural language processing (NLP) due to limited annotated resources and linguistic evolution spanning thousands of years. Addressing the challenges presented by this low annotated resource domain, we develop a relation extraction (RE) corpus that preserves the characteristics of classical Chinese documents. Utilizing this corpus, we explore RE in classical Chinese documents through a collaboration framework that integrates small pre-trained language models (SLMs), such as BERT, with large language models (LLMs) like GPT-3.5. SLMs can quickly adapt to specific tasks given sufficient supervised data but often struggle with few-shot scenarios. Conversely, LLMs leverage broad domain knowledge to handle few-shot challenges but face limitations when processing lengthy input sequences. Combining these complementary strengths, we propose a "train-guide-predict'' collaboration framework, where a small language model corporate with a large language model (SLCoLM). This framework enables SLMs to capture task-specific knowledge for head relation categories, while LLMs offer insights for few-shot relation categories. Experimental results show that SLCoLM outperforms both fine-tuned SLMs and LLMs using in-context learning (ICL). It also helps mitigate the long-tail problem in classical Chinese historical documents.

This paper explores how artificial intelligence is transforming engagement with cultural heritage from static preservation toward interactive knowledge production. Drawing on media-historical perspectives, it proposes the concept of knowledge liberation to interpret successive stages in the evolution of knowledge environments, from oral transmission and print culture to digital networks and AI-mediated interaction. Within this framework, cultural knowledge becomes progressively less constrained by the material conditions of its transmission. The paper examines several AI-enabled platforms developed at Peking University that support large-scale digitisation, structured data extraction, knowledge-graph construction, and multimodal cultural content generation. It argues that AI is emerging as a new knowledge medium that reshapes research methodologies, expands modes of cultural representation, and strengthens connections between humanities scholarship and public knowledge production.

Computational approaches to intertextuality have advanced from string matching to neural retrieval, yet their outputs, similarity scores and parallel-passage lists, identify where texts reuse one another without characterizing how or why. We recast fine-grained intertextuality extraction as an agentic task in which a large language model (LLM) reads two text units in full and, through a constrained tool interface, must ground each proposed reuse in exact character spans on both sides and label it under a five-dimension typology of reuse (form, aspect, source-marking, function, stance). We validate the approach on an exhaustive comparison of the Analects with the Book of Han, where three domain experts adjudicate a pooled multi-model candidate set into a benchmark of 2,533 intertextual pairs. Against this standard we study twelve LLMs, reporting precision (56

LLM-based research agents have advanced rapidly in science and engineering, where research is organized around executable experiments, code, and quantitative signals. Humanities scholarship, however, requires a different mode of reasoning: interpretive, evidence-grounded argument over primary sources, where scholarly value depends on faithful quotation, verifiable provenance, and close reading. Existing research agents remain largely optimized for execution and retrieval, not evidence-grounded interpretive reasoning. To address this gap, we introduce SPIRE (Scholarly-Primitives-Inspired Research Engine), a multi-agent framework for evidence-grounded humanities scholarship. Drawing on Scholarly Primitives theory, SPIRE casts recurring humanities operations as cooperating agent roles (source discovery, evidence annotation, comparison, provenance checking, sampling, citation binding, and argumentative synthesis) over a multi-scale close-reading substrate of passages, intra-context graph communities, and cross-context semantic clusters. On a peer-reviewed-paper benchmark over classical Chinese and Greco-Roman Latin scholarship, SPIRE recovers cited primary-source evidence more reliably than Naive LLM, Text RAG, and GraphRAG, and receives higher blind-judge scores on answer accuracy, depth, coverage, and evidence quality. Ablations show that both the scholarly-operation agents and close-reading retrieval contribute to evidence-grounded essays. Code, data catalogues, and reproduction scripts are released at https://github.com/YatingPan/SPIRE.

Existing scientific text representation methods based on contrastive learning typically adopt a binary classification paradigm of positive and negative samples, which struggles to capture the complex, hierarchical semantic similarity relationships inherent in scientific texts. To address this, we propose SciRep, a novel two-stage ranking distillation framework. In the first stage, we distill knowledge from a large language model to a medium-scale representation model using generated ranking samples; in the second stage, a multi-teacher strategy further transfers fine-grained ranking capability to a lightweight model. Evaluated on a scientific literature semantic embedding benchmark comprising three tasks, SciRep outperforms the strongest baseline by 11.3% in terms of Average Rank and also achieves the highest Mean Reciprocal Rank scores across all three tasks. These results demonstrate that the proposed ranking-aware distillation mechanism significantly enhances scientific text representation quality while maintaining efficient inference, offering a more effective contrastive learning method for domain-specific retrieval tasks.

The ancient Chinese texts exhibit marked intertextual characteristics, where scholars engage in citation, reinterpretation, and reconstruction of earlier works, forming an intellectual lineage spanning millennia. With advances in digital humanities, automated detection of text reuse in vast classical corpora has become feasible. However, existing algorithms remain largely confined to surface-level character matching, posing persistent challenges in identifying deep semantic correlations. To address this problem, we propose a novel text reuse detection method based on knowledge distillation for ancient Chinese literature which significantly enhances semantic understanding of classical texts while maintaining computational efficiency. Additionally, we construct a high-quality annotated dataset to establish a reliable benchmark for algorithmic evaluation. Through concrete case studies, we demonstrate the method’s applicability in cultural analysis, offering a novel technical pathway for the digitization and intelligent analysis of cultural heritage.

Generative AI (GenAI), as exemplified by ChatGPT, is transforming the way people seek information and interact with information systems and resources. This study investigates users' prompt formulation behavior through a longitudinal observation of experienced ChatGPT users. Extending prior research on prompt engineering, this study introduces the IIOC (Input‐Instruction‐Output‐Context) framework, delineating four core components: input, instruction, output, and context. The findings reveal that users have a strong preference for simple and direct prompt, with Single‐Component prompts accounting for 43.8% of all prompts. Dual‐Component combinations constitute 38.2%, with Input + Instruction (20.2%) being the most frequent pattern. Only 18.0% of prompts involve Multi‐Component combinations, indicating that complex prompt formulations are infrequent in typical user interactions. The findings also offer practical insights for user‐centered AI design by emphasizing the instruction, input, and output components that address users' core needs.

This work contributes to the digital humanities approach for studying premodern Chinese history and culture by creating a large-scale dataset annotated with named entities and relations. Through careful annotation guidelines and labeling of over 200,000 characters, we developed a dataset containing 30,000 named entities across six types and 7,000 relations spanning twenty categories. Experiments on named entity recognition (NER) using pre-trained language models and large language models on this dataset achieved an initial performance of NER (91.32 percent F1). In addition, relationship extraction (RE) on the pretrained language model achieves an 85.32 percent F1 score. While there is still room for improvement, our annotated dataset and models provide a useful starting point for extracting semantic information from premodern Chinese texts. It represents an effort to connect history and technology, increasing accessibility and preservation of premodern Chinese cultural treasures. Furthermore, our dataset can facilitate downstream tasks like culture analysis, knowledge graph construction, and computational understanding of premodern Chinese. Overall, this research represents a significant step toward digitally exploring premodern Chinese documents, providing a pathway for future work on knowledge organization and computational analysis of this valuable cultural legacy. Our code and data are available at: https://github.com/tangxuemei1995/AnChineseNERE

OBJECTIVE:1. Construct a risk prediction model to predict the factors of high intraoperative bleeding in patients undergoing posterior lumbar decompression and fusion internal fixation surgery during outpatient visits. 2. Implement pre-hospital blood management for surgery patients, to improve clinical outcomes. DESIGN & METHODS:We collected patients who underwent two-segment and three-segment posterior lumbar decompression and fusion internal fixation surgery in our hospital from 2016 to 2021. A total of 24 preoperative indicators were analyzed, covering medical history, demographic characteristics, segment, operator and laboratory test results. We used a logistic regression model to optimize the model's feature selection. The predictive model was constructed using the multivariable logistic regression method with all included methods, and a nomogram was created to display the model. Activated partial thromboplastin time, surgeon volume, American Society of Anesthesiologists classification, body mass index, and the number of fusion and fixation lumbar segments were used to construct the predictive model. The predictive model's discrimination, calibration, clinical applicability, and rationality were evaluated. RESULTS:The predictive model's area under the receiver operating characteristic curve is 0.723, with a 95% confidence interval of (0.685-0.760). The training set's decision curve analysis demonstrates that applying this diagnostic curve will increase the net benefit when the threshold probability is between 5% and 40%. CONCLUSION:This study developed a novel nomogram with relatively good accuracy to assist clinical doctors in assessing the high intraoperative bleeding risk in patients undergoing posterior lumbar decompression and fusion internal fixation surgery during outpatient visits. By evaluating individual risk, surgeons can develop an individualized treatment plan to reduce the risk of intraoperative bleeding for each patient.

Recent advances in unsupervised sentence representation learning have relied heavily on contrastive objectives over pre-trained language models (PLMs). However, the quality of learned representations is often limited by shallow layers and restricted model capacity, especially in the absence of large-scale supervision. To address this, we propose SageRep, a novel layer-adaptive self-knowledge distillation framework that improves representation quality without requiring external teacher models or multi-stage training. Motivated by the observation that deeper layers in PLMs encode richer semantic signals, SageRep enables shallower layers to distill knowledge dynamically from deeper ones, guided by sentence-level similarity distributions. Unlike prior static self-knowledge distillation methods, our approach adaptively selects the most informative teacher layers per instance, promoting more effective intra-model knowledge transfer. Additionally, we introduce a contrastive regularization strategy using inter-layer negatives to mitigate representation over-smoothing. Extensive evaluations on standard semantic textual similarity benchmarks demonstrate that SageRep achieves superior performance over previous unsupervised methods, with minimal additional training cost.

Some argue that the essence of humanity, such as creativity and sentiment, can never be mimicked by machines. This paper casts doubt on this belief by studying a vital question: Can AI compose poetry as well as humans? To answer the question, we propose ProFTAP, a novel evaluation framework inspired by Turing test to assess AI's poetry writing capability. We apply it on current large language models (LLMs) and find that recent LLMs do indeed possess the ability to write classical Chinese poems nearly indistinguishable from those of humans. We also reveal that various open-source LLMs can outperform GPT-4 on this task.

Chinese ancient book catalogs are important cultural heritage and academic resources for the study of ancient Chinese history and culture. These catalogs need to be curated so that their value can be fully exploited in today's digital environment. This study is based on a collaborative curation project where eight representative ancient catalogs were curated into a diachronic dataset and tools to discover and analyze the data were developed. We reviewed literature and consulted humanities scholars to derive the characteristics and curation requirements of the ancient catalogs. A collaborative model was proposed based on the requirements to guide the curation process. This model reveals the duality of humanities scholars' role in collaborative curation and depicts main curation activities including metadata and description, appraisal and selection, data processing, developing tools, access and use, and evolution. Lessons learned from the curation practice include two main issues-project personnel and humanities scholars' acceptance of visualization. The study also yields a dataset and a set of tools that can be directly used by scholars interested in knowledge organization and ancient catalog related topics.

Text line detection is a key task in historical document analysis facing many challenges of arbitrary-shaped text lines, dense texts, and text lines with high aspect ratios, etc. In this paper, we propose a general framework for historical document text detection (SegHist), enabling existing segmentation-based text detection methods to effectively address the challenges, especially text lines with high aspect ratios. Integrating the SegHist framework with the commonly used method DB++, we develop DB-SegHist. This approach achieves state-of-the-art (SOTA) on the IACC2022CHDAC (CHDAC), MTHv2, and competitive results on ICDAR2019HDRC Chinese (HDRC) datasets, with a significant improvement of 1.19

Introduction. Much of humanities data are existing materials. Yet there are few works examining humanities scholars’ data seeking behaviours. The study aims to address this gap building upon Ellis’ model. Method. We recruited 27 humanities scholars with diverse academic backgrounds and conducted in-depth interviews. Analysis. A preliminary codebook was developed from existing literature. Researchers combined deductive and inductive coding to analyse the interview transcripts. Results. Humanities scholars’ data interactions fall into two approaches: data-driven and structure-driven, each involving three phases – exploratory seeking, focused seeking, and supplementary seeking. We identified eleven characteristics of data seeking behaviours operating at different levels and revealed their variations across research approaches and seeking phases. Conclusion. The study contributes to the conceptual growth of Ellis' model and expands its utility beyond the original information seeking contexts, indicating its potential applicability to data seeking. It also provides practical implications for system design and humanities data curation.

Natural Language Processing (NLP) plays a pivotal role in the realm of Digital Humanities (DH) and serves as the cornerstone for advancing the structural analysis of historical and cultural heritage texts. This is particularly true for the domains of named entity recognition (NER) and relation extraction (RE). In our commitment to expediting ancient history and culture, we present the “Chinese Historical Information Extraction Corpus”(CHisIEC). CHisIEC is a meticulously curated dataset designed to develop and evaluate NER and RE tasks, offering a resource to facilitate research in the field. Spanning a remarkable historical timeline encompassing data from 13 dynasties spanning over 1830 years, CHisIEC epitomizes the extensive temporal range and text heterogeneity inherent in Chinese historical documents. The dataset encompasses four distinct entity types and twelve relation types, resulting in a meticulously labeled dataset comprising 14,194 entities and 8,609 relations. To establish the robustness and versatility of our dataset, we have undertaken comprehensive experimentation involving models of various sizes and paradigms. Additionally, we have evaluated the capabilities of Large Language Models (LLMs) in the context of tasks related to ancient Chinese history. The dataset and code are available at .

Cultural heritage serves as the enduring record of human thought and history. Despite significant efforts dedicated to the preservation of cultural relics, many ancient artefacts have been ravaged irreversibly by natural deterioration and human actions. Deep learning technology has emerged as a valuable tool for restoring various kinds of cultural heritages, including ancient text restoration. Previous research has approached ancient text restoration from either visual or textual perspectives, often overlooking the potential of synergizing multimodal information. This paper proposes a novel Multimodal Multitask Restoring Model (MMRM) to restore ancient texts, particularly emphasising the ideograph. This model combines context understanding with residual visual information from damaged ancient artefacts, enabling it to predict damaged characters and generate restored images simultaneously. We tested the MMRM model through experiments conducted on both simulated datasets and authentic ancient inscriptions. The results show that the proposed method gives insightful restoration suggestions in both simulation experiments and real-world scenarios. To the best of our knowledge, this work represents the pioneering application of multimodal deep learning in ancient text restoration, which will contribute to the understanding of ancient society and culture in digital humanities fields.

PurposeRecent trends have shown the integration of Chinese word segmentation (CWS) and part-of-speech (POS) tagging to enhance syntactic and semantic parsing. However, the potential utility of hierarchical and structural information in these tasks remains underexplored. This study aims to leverage multiple external knowledge sources (e.g. syntactic and semantic features, lexicons) through various modules for the joint task.Design/methodology/approachWe introduce a novel learning framework for the joint CWS and POS tagging task, utilizing graph convolutional networks (GCNs) to encode syntactic structure and semantic features. The framework also incorporates a pre-defined lexicon through a lexicon attention module. We evaluate our model on a range of public corpora, including CTB5, PKU and UD, the novel ZX dataset and the comprehensive CTB9 dataset.FindingsExperimental results on these benchmark corpora demonstrate the effectiveness of our model in improving the performance of the joint task. Notably, we find that syntax information significantly enhances performance, while lexicon information helps mitigate the issue of out-of-vocabulary (OOV) words.Originality/valueThis study introduces a comprehensive approach to the joint CWS and POS tagging task by combining multiple features. Moreover, the proposed framework offers potential adaptability to other sequence labeling tasks, such as named entity recognition (NER).

Quantitative cultural studies have witnessed a surge with the rapid development of computer technology in recent years. Since ancient literature constitutes a long-time-span repository for human culture, with quantitative methods and ancient texts, scholars can study the genesis and progression of human history and society across historical epochs from digital perspectives. Nevertheless, traditional humanities scholars often lack the requisite technical skills, creating a demand for interactive platforms. This paper introduces the Evol platform—an online tool designed for the quantitative analysis of ancient literature. Equipped with various analysis functions and visualization tools, the Evol platform allows users to quantify literary documents through intuitive online interaction. Using this platform, we investigated three cases of cultural evolution in ancient Chinese history: (1) the changing attitude of the government towards nomadic ethnic groups; (2) the formulation and propagation of an allusion phrase related to the Battle of Muye; (3) the influence of the Book of Changes across diverse cultural domains. By showcasing cases across diverse semantic units and topics, Evol demonstrates its potential in providing efficient and low-cost experimental tools catering to the realms of culturomics, history, and philology.

Recently, deep learning methods have achieved remarkable success in the Chinese word segmentation (CWS) task. Some of them enhance the CWS model by utilizing contextual features and external resources (e.g., sub-words, lexicon, and syntax). However, existing approaches fail to fully use the heterogeneous features and their structural information. Therefore, in this paper, we propose a heterogeneous information learning framework for CWS, named heterogeneous graph neural segmenter (HGNSeg), which exploits heterogeneous features with the graph convolutional networks and the pretrained language model. Experimental results on six benchmark datasets (e.g., SIGHAN 2005 and SIGHAN 2008) confirm that HGNSeg can effectively improve the performance of CWS. Importantly, HGNSeg also demonstrates an excellent ability to alleviate the out-of-vocabulary (OOV) issue in cross-domain scenarios.

Recently, large language models (LLMs) have been successful in relational extraction (RE) tasks, especially in the few-shot learning. An important problem in the field of RE is long-tailed data, while not much attention is paid to this problem using LLM approaches. Therefore, in this paper, we propose SLCoLM, a model collaboration framework, to mitigate the data long-tail problem. In our framework, we use the “Training-Guide-Predict” strategy to combine the strengths of small pre-trained language models (SLMs) and LLMs, where a task-specific SLM framework acts as a guider, transfers task knowledge to the LLM and guides the LLM in performing RE tasks. Our experiments on an ancient Chinese RE dataset rich in relation types show that the approach facilitates RE of long-tail relation types.