Text-to-SQL translates natural language questions into executable SQL queries, enabling intuitive database access for non-experts. While large language models achieve strong performance on Text-to-SQL with prompting, they still struggle with complex queries that involve deeply nested logic or multiple clauses. A widely used approach employs SQL skeletons--intermediate representations of query logic--to streamline generation, but existing methods are limited by their reliance on a single structural hypothesis and lack of progressive reasoning. To overcome these limitations, we propose LEAF-SQL, a novel framework that reframes skeleton prediction as a coarse-to-fine tree search process. LEAF-SQL enables systematic exploration of diverse structural hypotheses with adaptive refinement. Several key techniques are employed in LEAF-SQL: (1) a three-level skeleton hierarchy to guide the search, (2) a Skeleton Formulation Agent to generate diverse candidates, and (3) a Skeleton Evaluation Agent to efficiently prune the search space. This integrated design yields skeleton candidates that are both structurally diverse and granularity-adaptive, providing a stronger foundation for the SQL generation. Extensive experiments show that LEAF-SQL consistently improves the performance of various LLM backbones. On the official hidden test set of the challenging BIRD benchmark, our method achieves 71.6 execution accuracy, which outperforms leading search-based and skeleton-based methods, affirming its effectiveness for complex queries.

Recent advancements in event argument extraction (EAE) involve incorporating useful auxiliary information into models during training and inference, such as retrieved instances and event templates. These methods face two challenges: (1) the retrieval results may be irrelevant and (2) templates are developed independently for each event without considering their possible relationship. In this work, we propose DEGAP to address these challenges through a simple yet effective components: dual prefixes, i.e. learnable prompt vectors, where the instance-oriented prefix and template-oriented prefix are trained to learn information from different event instances and templates. Additionally, we propose an event-guided adaptive gating mechanism, which can adaptively leverage possible connections between different events and thus capture relevant information from the prefix. Finally, these event-guided prefixes provide relevant information as cues to EAE model without retrieval. Extensive experiments demonstrate that our method achieves new state-of-the-art performance on four datasets (ACE05, RAMS, WIKIEVENTS, and MLEE). Further analysis shows the impact of different components.

Event extraction is a long-standing and challenging task in natural language processing, and existing studies mainly focus on extracting events within sentences. However, a significant problem that has not been carefully investigated is whether an “event topic” can be identified to represent the main aspects of extracted events. This article formulates the “topic event” extraction problem, aiming to identify a representative event from extracted ones. Specifically, after defining the topic event, we develop a multifocal graph-based framework to handle the extraction task. To enrich the associations of events and their tokens, we construct four event graphs, including the event subgraph and three event-associated graphs (i.e., event dependency parsing graph, event organization graph, and event share token graph), that reflect the internal and external structures of events, respectively. Subsequently, we design a multi-attention event-graph neural network to capture these event graph structures and improve event subgraph embedding. Finally, the output embeddings in the last layer of each channel are concatenated and fed into a fully connected network for topic event recognition. Extensive experiments validate the effectiveness of our method, and the results confirm its superiority over state-of-the-art baselines. In-depth analyses explore the essential factors (e.g., graph structures, attentions, feature generation method, etc.) determining the extraction performance.

Existing models on event detection share three-fold limitations, including (1) insufficient consideration of the structures between dependency relations, (2) limited exploration of the directed-edge semantics, and (3) issues in strengthening the event core arguments. To tackle these problems, we propose a dependency structure-enhanced event detection framework. In addition to the traditional token dependency parsing tree, denoted as TDG, our model considers the dependency edges in it as new nodes and constructs a dependency relation graph (DRG). DRG allows the embedding representations of dependency relations to be updated as nodes rather than edges in a graph neural network. Moreover, the levels of core argument nodes in the two graphs are adjusted by dependency relation types in TDG to enhance their status. Subsequently, the two graphs are further encoded and jointly trained in graph attention networks (GAT). Importantly, we design an interaction strategy of node embedding for the two graphs and refine the attention coefficient computational method to encode the semantic meaning of directed edges. Extensive experiments are conducted to validate the effectiveness of our method, and the results confirm its superiority over the state-of-the-art baselines. Our model outperforms the best benchmark with the F1 score increased by 3.5 and 3.4 percentage points on ACE2005 English and Chinese corpus.

Large language models (LLMs) with prompting have achieved encouraging results on many natural language processing (NLP) tasks based on task-tailored promptings. Text-to-SQL is a critical task that generates SQL queries from natural language questions. However, prompting on LLMs haven’t show superior performance on Text-to-SQL task due to the absence of tailored promptings. In this work, we propose three promptings specifically designed for Text-to-SQL: SL-prompt, CC-prompt, and SL+CC prompt. SL-prompt is designed to guide LLMs to identify relevant tables; CC-prompt directs LLMs to generate SQL clause by clause; and SL+CC prompt is proposed to combine the strengths of these above promptings. The three prompting strategies makes three solutions for Text-to-SQL. Then, another prompting strategy, the RS-prompt is proposed to direct LLMs to select the best answer from the results of the solutions. We conducted extensive experiments, and experimental results show that our method achieved an execution accuracy of 86.2% and a test-suite accuracy of 76.9%, which is 1.1%, and 2.7% higher than the current state-of-the-art Text-to-SQL methods, respectively. The results confirmed that the proposed promptings enhanced the capabilities of LLMs on Text-to-SQL. Experimental results also show that the granularity of schema linking and the order of clause generation have great impact on the performance, which are considered little in previous research.

Document-level event argument extraction aims to identify argument spans and predict the roles they play in the event from a single document. To enhance argument extraction by better leveraging relevant context information, most existing methods focus solely on context information from the separate perspective of either event or role, without concurrently considering the influence of both event and role. Additionally, existing methods struggle to effectively handle the many-to-many relationship between arguments and roles. Therefore, this paper introduces the CasDEE (Cascade Constrained Pooling and Relevance Evaluation for Document-level Event Argument Extraction) model, which filters context information relevant to the target event and its roles following a constraining pathway of event, role, and argument boundary, and integrates it into the representations of both the candidate argument and the role. To tackle the many-to-many relationship between arguments and roles, CasDEE first utilizes a relevance function to evaluate the relevance of each span-role pair, and then selects all span-role pairs whose relevance are greater than the given threshold as the final argument extraction result. Extensive experiments show that CasDEE achieves state-of-the-art performance on RAMS and WikiEvents datasets. Further experiments reveal the superior performance of CasDEE in addressing long-range dependencies, handling the many-to-many relationship between arguments and roles, and multiple events extraction.

Document-level event extraction is a long-standing challenging information retrieval problem involving a sequence of sub-tasks: entity extraction, event type judgment, and event type-specific multi-event extraction. However, addressing the problem as multiple learning tasks leads to increased model complexity. Also, existing methods insufficiently utilize the correlation of entities crossing different events, resulting in limited event extraction performance. This article introduces a novel framework for document-level event extraction, incorporating a new data structure called token-event-role and a multi-channel argument role prediction module. The proposed data structure enables our model to uncover the primary role of tokens in multiple events, facilitating a more comprehensive understanding of event relationships. By leveraging the multi-channel prediction module, we transform entity and multi-event extraction into a single task of predicting token–event pairs, thereby reducing the overall parameter size and enhancing model efficiency. The results demonstrate that our approach outperforms the state-of-the-art method by 9.5 percentage points in terms of the F 1 score, highlighting its superior performance in event extraction. Furthermore, an ablation study confirms the significant value of the proposed data structure in improving event extraction tasks, further validating its importance in enhancing the overall performance of the framework.

Learning topic hierarchies from a multi-domain corpus is crucial in topic modeling as it reveals valuable structural information embedded within documents. Despite the extensive literature on hierarchical topic models, effectively discovering inter-topic correlations and differences among subtopics at the same level in the topic hierarchy, obtained from multiple domains, remains an unresolved challenge. This article proposes an enhanced nested Chinese restaurant process (nCRP), nCRP+, by introducing an additional mechanism based on Chinese restaurant franchise (CRF) for aspect-sharing pattern extraction in the original nCRP. Subsequently, by employing the distribution extracted from nCRP+ as the prior distribution for topic hierarchy in the hierarchical Dirichlet processes (HDP), we develop a hierarchical topic model for multi-domain corpus, named rHDP. We describe the model with the analogy of Chinese restaurant franchise based on the central kitchen and propose a hierarchical Gibbs sampling scheme to infer the model. Our method effectively constructs well-established topic hierarchies, accurately reflecting diverse parent-child topic relationships, explicit topic aspect sharing correlations for inter-topics, and differences between these shared topics. To validate the efficacy of our approach, we conduct experiments using a renowned public dataset and an online collection of Chinese financial documents. The experimental results confirm the superiority of our method over the state-of-the-art techniques in identifying multi-domain topic hierarchies, according to multiple evaluation metrics.

Qizhi Wan, Changxuan Wan, Keli Xiao, Dexi Liu, Chenliang Li, Bolong Zheng, Xiping Liu, Rong Hu. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Event extraction is an essential task in natural language processing. Although extensively studied, existing work shares issues in three aspects, including (1) the limitations of using original syntactic dependency structure, (2) insufficient consideration of the node level and type information in Graph Attention Network (GAT), and (3) insufficient joint exploitation of the node dependency type and part-of-speech (POS) encoding on the graph structure. To address these issues, we propose a novel framework for open event extraction in documents. Specifically, to obtain an enhanced dependency structure with powerful encoding ability, our model is capable of handling an enriched parallel structure with connected ellipsis nodes. Moreover, through a bidirectional dependency parsing graph, it considers the sequence of order structure and associates the ancestor and descendant nodes. Subsequently, we further exploit node information, such as the node level and type, to strengthen the aggregation of node features in our GAT. Finally, based on the coordination of triple-channel features (i.e., semantic, syntactic dependency and POS), the performance of event extraction is significantly improved. Extensive experiments are conducted to validate the effectiveness of our method, and the results confirm its superiority over the state-of-the-art baselines. Furthermore, in-depth analyses are provided to explore the essential factors determining the extraction performance.

We construct a Chinese Economic Event Treebank (CEETB) , focusing on revealing economic and finance events and their relations. Investigating economic event relations will benefit academic research and practice in not just economics but many other scientific areas. The characteristics of economic-related texts (e.g., abundant longer enterprises names and terms) and the Chinese language speciality (e.g., component ellipsis in long sentences) have resulted in challenges in the event relation extraction task. Existing Chinese corpora containing economic event relations mainly focused on finance areas (e.g., the equity market) and only covered a few event types. To support research that may involve economic text analysis in Chinese, our CEETB is constructed following a carefully designed process. First, based on practical and research requirements, we summarize nine different types of event relations and four types of component ellipses in economic texts. Then, an excellent annotation scheme is presented to hyalinize the model, strategy, and process in annotation, followed by statistical analysis and quality evaluation for the CEETB corpus. Finally, to demonstrate the strengths of the constructed corpus in practical applications, we conduct experiments on five SOTA models for event relation extraction.