We present AutoSchemaKG, a framework for fully autonomous knowledge graph construction that eliminates the need for predefined schemas. Our system leverages large language models to simultaneously extract knowledge triples and induce comprehensive schemas directly from text, modeling both entities and events while employing conceptualization to organize instances into semantic categories. Processing over 50 million documents, we construct ATLAS (Automated Triple Linking And Schema induction), a family of knowledge graphs with 900+ million nodes and 5.9 billion edges. This approach outperforms state-of-the-art baselines on multi-hop QA tasks and enhances LLM factuality. Notably, our schema induction achieves 92\% semantic alignment with human-crafted schemas with zero manual intervention, demonstrating that billion-scale knowledge graphs with dynamically induced schemas can effectively complement parametric knowledge in large language models.

Hong Ting Tsang, Jiaxin Bai, Haoyu Huang, Qiao Xiao, Tianshi Zheng, Baixuan Xu, Shujie Liu, Yangqiu Song. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Deductive and abductive reasoning are two critical paradigms for analyzing knowledge graphs, enabling applications from financial query answering to scientific discovery. Deductive reasoning on knowledge graphs usually involves retrieving entities that satisfy a complex logical query, while abductive reasoning generates plausible logical hypotheses from observations. Despite their clear synergistic potential, where deduction can validate hypotheses and abduction can uncover deeper logical patterns, existing methods address them in isolation. To bridge this gap, we propose DARK, a unified framework for Deductive and Abductive Reasoning in Knowledge graphs. As a masked diffusion model capable of capturing the bidirectional relationship between queries and conclusions, DARK has two key innovations. First, to better leverage deduction for hypothesis refinement during abductive reasoning, we introduce a self-reflective denoising process that iteratively generates and validates candidate hypotheses against the observed conclusion. Second, to discover richer logical associations, we propose a logic-exploration reinforcement learning approach that simultaneously masks queries and conclusions, enabling the model to explore novel reasoning compositions. Extensive experiments on multiple benchmark knowledge graphs show that DARK achieves competitive performance on both deductive and abductive reasoning tasks, demonstrating the significant benefits of our unified approach.

Weiqi Wang, Limeng Cui, Xin Liu, Sreyashi Nag, Wenju Xu, Chen Luo, Sheikh Muhammad Sarwar, Yang Li, Hansu Gu, Hui Liu, Changlong Yu, Jiaxin Bai, Yifan Gao, Haiyang Zhang, Qi He, Shuiwang Ji, Yangqiu Song. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Tianshi Zheng, Jiazheng Wang, Zihao Wang, Jiaxin Bai, Hang Yin, Zheye Deng, Yangqiu Song, Jianxin Li. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Large Language Models (LLMs) are catalyzing a paradigm shift in scientific discovery, evolving from task-specific automation tools into increasingly autonomous agents and fundamentally redefining research processes and human-AI collaboration.This survey systematically charts this burgeoning field, placing a central focus on the changing roles and escalating capabilities of LLMs in science.Through the lens of the scientific method, we introduce a foundational three-level taxonomy-Tool, Analyst, and Scientist-to delineate their escalating autonomy and evolving responsibilities within the research lifecycle.We further identify pivotal challenges and future research trajectories such as robotic automation, self-improvement, and ethical governance.Overall, this survey provides a conceptual architecture and strategic foresight to navigate and shape the future of AI-driven scientific discovery, fostering both rapid innovation and responsible advancement.

Tianshi Zheng, Cheng Jiayang, Chunyang Li, Haochen Shi, Zihao Wang, Jiaxin Bai, Yangqiu Song, Ginny Wong, Simon See. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

The existing research on hybrid energy storage system (HESS) stability mostly focuses on small-signal stability analysis methods. However, when large-signal disturbances occur, the intrinsic nonlinearity of the HESS becomes inevitable so that the small-signal studies are no longer valid, which is beyond the scope of small-signal stability analysis. To address this problem, a large-signal stability analysis method of the HESS is proposed based on the estimation of domain of attraction (DOA). This method can not only analyze the asymptotic stability of the operating point, but also answer the question of how large deviations from the operating point can be tolerated by the system. Specifically, the state space model and Takagi-Sugeno (T-S) fuzzy model of the whole system are first constructed. By using linear matrix inequality (LMI) and Lyapunov’s method, the DOA of the system at the operating point is estimated, and the large-signal stability of HESS is analyzed. On this basis, the influence of system parameters (e.g. virtual resistance, virtual inductance and PI regulators) and load power on the large-signal stability are studied, and the dominant parameters affecting the large-signal stability of the HESS is clarified, which provides a practical guiding basis for the parameter optimization design of the HESS to ensure the safe and stable operation of the system under extreme load switching conditions. The simulation results verify the feasibility and effectiveness of the proposed large-signal stability analysis method and the correctness of the analyzing results.

In the more electric aircraft (MEA) power system, the voltage stabilization of the interface three-phase rectifier is difficult to be achieved due to the wide range varying frequency and the unbalanced amplitude and phase of the output voltage of the AC generator. In this article, an advanced voltage control strategy based on the techniques of the dynamic phase voltages reconstruction and the improved proportional resonant (PR) regulation is proposed. Thanks to the proposed dynamic phase voltages reconstruction algorithm, the phase-locked loop (PLL) is not required, avoiding the risk of PLL instability in the presence of wide range frequency fluctuations. By integrating the improved PR regulator, the error steady-state control of the currents in αβ frame is guaranteed and the negative sequence harmonics are suppressed, which greatly reduces the reactive power and, hence, improves the power factor. Some simulations have been carried out to validate the effectiveness of the proposed strategy.

This paper addresses the challenge of insufficient sampling diversity in large language models (LLMs) under the conventional autoregressive decoding framework. Our research reveals that, in some tasks where LLMs underperform, they actually possess the capability to provide correct answers. However, this potential is limited by inadequate sampling diversity in the decoding process, which prevents the models from effectively and fully leveraging their reasoning capabilities. To address this issue, we propose the Adaptive Trigram Model-Assisted Decoding Strategy (ATM-ADS), a method designed to enhance the reasoning abilities of large models without requiring fine-tuning. By dynamically providing diverse candidate vocabularies, combined with a decoding dynamic decision-making mechanism and an adaptive smoothing strategy, our approach substantially enhances sampling diversity during the decoding process. Experimental results demonstrate that our ATM-ADS significantly improves model performance across multiple task scenarios, highlighting its broad potential for cross-task and cross-linguistic applications.

Weiqi Wang, Tianqing Fang, Chunyang Li, Haochen Shi, Wenxuan Ding, Baixuan Xu, Zhaowei Wang, Jiaxin Bai, Xin Liu, Cheng Jiayang, Chunkit Chan, Yangqiu Song. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Abductive reasoning is the process of making educated guesses to provide explanations for observations.Although many applications require the use of knowledge for explanations, the utilization of abductive reasoning in conjunction with structured knowledge, such as a knowledge graph, remains largely unexplored.To fill this gap, this paper introduces the task of complex logical hypothesis generation, as an initial step towards abductive logical reasoning with KG.In this task, we aim to generate a complex logical hypothesis so that it can explain a set of observations.We find that the supervised trained generative model can generate logical hypotheses that are structurally closer to the reference hypothesis.However, when generalized to unseen observations, this training objective does not guarantee better hypothesis generation.To address this, we introduce the Reinforcement Learning from Knowledge Graph (RLF-KG) method, which minimizes differences between observations and conclusions drawn from generated hypotheses according to the KG.Experiments show that, with RLF-KG's assistance, the generated hypotheses provide better explanations, and achieve stateof-the-art results on three widely used KGs. 1

Baixuan Xu, Weiqi Wang, Haochen Shi, Wenxuan Ding, Huihao Jing, Tianqing Fang, Jiaxin Bai, Xin Liu, Changlong Yu, Zheng Li, Chen Luo, Qingyu Yin, Bing Yin, Long Chen, Yangqiu Song. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

Classification of hyperspectral images (HSIs) is a crucial topic in the domain of remote sensing. However, existing HSI classification methods often fail to adequately consider the connection between mid-level and high-level spectral-spatial features. Consequently, we propose a novel method named generative adversarial network-based spectral–spatial learning (GAN-SSL) for HSI classification. The method leverages the high spatial resolution of panchromatic (PAN) images and combines PAN images with HSIs to generate higher-quality HSIs, resulting in improved classification accuracy. Firstly, the dual-stream GAN is employed to generate higher-quality HSI from both HSI and PAN images. Secondly, the shallow-deep feature extraction classification network is utilized to capture both mid-level and high-level spectral–spatial information from the obtained features and a classifier is employed to categorize the extracted information. The experimental results obtained from two datasets indicate that the proposed approach surpasses several state-of-the-art techniques in terms of classification performance.

Understanding user intentions is essential for improving product recommendations, navigation suggestions, and query reformulations. However, user intentions can be intricate, involving multiple sessions and attribute requirements connected by logical operators such as And, Or, and Not. For instance, a user may search for Nike or Adidas running shoes across various sessions, with a preference for purple. In another example, a user may have purchased a mattress in a previous session and is now looking for a matching bed frame without intending to buy another mattress. Existing research on session understanding has not adequately addressed making product or attribute recommendations for such complex intentions. In this paper, we present the task of logical session complex query answering (LS-CQA), where sessions are treated as hyperedges of items, and we frame the problem of complex intention understanding as an LS-CQA task on an aggregated hypergraph of sessions, items, and attributes. This is a unique complex query answering task with sessions as ordered hyperedges. We also introduce a new model, the Logical Session Graph Transformer (LSGT), which captures interactions among items across different sessions and their logical connections using a transformer structure. We analyze the expressiveness of LSGT and prove the permutation invariance of the inputs for the logical operators. By evaluating LSGT on three datasets, we demonstrate that it achieves state-of-the-art results.

In the era of large language models (LLMs), efficient and accurate data retrieval has become increasingly crucial for the use of domain-specific or private data in the retrieval augmented generation (RAG). Neural graph databases (NGDBs) have emerged as a powerful paradigm that combines the strengths of graph databases (GDBs) and neural networks to enable efficient storage, retrieval, and analysis of graph-structured data which can be adaptively trained with LLMs. The usage of neural embedding storage and Complex neural logical Query Answering (CQA) provides NGDBs with generalization ability. When the graph is incomplete, by extracting latent patterns and representations, neural graph databases can fill gaps in the graph structure, revealing hidden relationships and enabling accurate query answering. Nevertheless, this capability comes with inherent trade-offs, as it introduces additional privacy risks to the domain-specific or private databases. Malicious attackers can infer more sensitive information in the database using well-designed queries such as from the answer sets of where Turing Award winners born before 1950 and after 1940 lived, the living places of Turing Award winner Hinton are probably exposed, although the living places may have been deleted in the training stage due to the privacy concerns. In this work, we propose a privacy-preserved neural graph database (P-NGDB) framework to alleviate the risks of privacy leakage in NGDBs. We introduce adversarial training techniques in the training stage to enforce the NGDBs to generate indistinguishable answers when queried with private information, enhancing the difficulty of inferring sensitive information through combinations of multiple innocuous queries. Extensive experimental results on three datasets show that our framework can effectively protect private information in the graph database while delivering high-quality public answers responses to queries. The code is available at https://github.com/HKUST-KnowComp/PrivateNGDB.

A complex logic query in a knowledge graph refers to a query expressed in logic form that conveys a complex meaning, such as where did the Canadian Turing award winner graduate from? Knowledge graph reasoning-based applications, such as dialogue systems and interactive search engines, rely on the ability to answer complex logic queries as a fundamental task. In most knowledge graphs, edges are typically used to either describe the relationships between entities or their associated attribute values. An attribute value can be in categorical or numerical format, such as dates, years, sizes, etc. However, existing complex query answering (CQA) methods simply treat numerical values in the same way as they treat entities. This can lead to difficulties in answering certain queries, such as which Australian Pulitzer award winner is born before 1927, and which drug is a pain reliever and has fewer side effects than Paracetamol. In this work, inspired by the recent advances in numerical encoding and knowledge graph reasoning, we propose numerical complex query answering. In this task, we introduce new numerical variables and operations to describe queries involving numerical attribute values. To address the difference between entities and numerical values, we also propose the framework of Number Reasoning Network (NRN) for alternatively encoding entities and numerical values into separate encoding structures. During the numerical encoding process, NRN employs a parameterized density function to encode the distribution of numerical values. During the entity encoding process, NRN uses established query encoding methods for the original CQA problem. Experimental results show that NRN consistently improves various query encoding methods on three different knowledge graphs and achieves state-of-the-art results.

Querying knowledge graphs (KGs) using deep learning approaches can naturally leverage the reasoning and generalization ability to learn to infer better answers. Traditional neural complex query answering (CQA) approaches mostly work on entity-centric KGs. However, in the real world, we also need to make logical inferences about events, states, and activities (i.e., eventualities or situations) to push learning systems from System I to System II, as proposed by Yoshua Bengio. Querying logically from an EVentuality-centric KG (EVKG) can naturally provide references to such kind of intuitive and logical inference. Thus, in this paper, we propose a new framework to leverage neural methods to answer complex logical queries based on an EVKG, which can satisfy not only traditional first-order logic constraints but also implicit logical constraints over eventualities concerning their occurrences and orders. For instance, if we know that *Food is bad* happens before *PersonX adds soy sauce*, then *PersonX adds soy sauce* is unlikely to be the cause of *Food is bad* due to implicit temporal constraint. To facilitate consistent reasoning on EVKGs, we propose Complex Eventuality Query Answering (CEQA), a more rigorous definition of CQA that considers the implicit logical constraints governing the temporal order and occurrence of eventualities. In this manner, we propose to leverage theorem provers for constructing benchmark datasets to ensure the answers satisfy implicit logical constraints. We also propose a Memory-Enhanced Query Encoding (MEQE) approach to significantly improve the performance of state-of-the-art neural query encoders on the CEQA task.

Various data mining tasks have been proposed to study Community Question Answering (CQA) platforms like Stack Overflow. The relatedness between some of these tasks provides useful learning signals to each other via Multi-Task Learning (MTL). However, due to the high heterogeneity of these tasks, few existing works manage to jointly solve them in a unified framework. To tackle this challenge, we develop a multi-relational graph based MTL model called Heterogeneous Multi-Task Graph Isomorphism Network (HMTGIN) which efficiently solves heterogeneous CQA tasks. In each training forward pass, HMTGIN embeds the input CQA forum graph by an extension of Graph Isomorphism Network and skip connections. The embeddings are then shared across all task-specific output layers to compute respective losses. Moreover, two cross-task constraints based on the domain knowledge about tasks' relationships are used to regularize the joint learning. In the evaluation, the embeddings are shared among different task-specific output layers to make corresponding predictions. To the best of our knowledge, HMTGIN is the first MTL model capable of tackling CQA tasks from the aspect of multi-relational graphs. To evaluate HMTGIN's effectiveness, we build a novel large-scale multi-relational graph CQA dataset with over two million nodes from Stack Overflow. Extensive experiments show that: (1) HMTGIN is superior to all baselines on five tasks; (2) The proposed MTL strategy and cross-task constraints have substantial advantages.

Character linking, the task of linking mentioned people in conversations to the real world, is crucial for understanding the conversations.For the efficiency of communication, humans often choose to use pronouns (e.g., "she") or normal phrases (e.g., "that girl") rather than named entities (e.g., "Rachel") in the spoken language, which makes linking those mentions to real people a much more challenging than a regular entity linking task.To address this challenge, we propose to incorporate the richer context from the coreference relations among different mentions to help the linking.On the other hand, considering that finding coreference clusters itself is not a trivial task and could benefit from the global character information, we propose to jointly solve these two tasks.Specifically, we propose C 2 , the joint learning model of Coreference resolution and Character linking.The experimental results demonstrate that C 2 can significantly outperform previous works on both tasks.Further analyses are conducted to analyze the contribution of all modules in the proposed model and the effect of all hyper-parameters.

Quantitative Investment, built on the solid foundation of robust financial theories, is at the center stage in investment industry today. The essence of quantitative investment is the multi-factor model, which explains the relationship between the risk and return of equities. However, the multi-factor model generates enormous quantities of factor data, through which even experienced portfolio managers find it difficult to navigate. This has led to portfolio analysis and factor research being limited by a lack of intuitive visual analytics tools. Previous portfolio visualization systems have mainly focused on the relationship between the portfolio return and stock holdings, which is insufficient for making actionable insights or understanding market trends. In this paper, we present s Portfolio, which, to the best of our knowledge, is the first visualization that attempts to explore the factor investment area. In particular, sPortfolio provides a holistic overview of the factor data and aims to facilitate the analysis at three different levels: a Risk-Factor level, for a general market situation analysis; a Multiple-Portfolio level, for understanding the portfolio strategies; and a Single-Portfolio level, for investigating detailed operations. The system's effectiveness and usability are demonstrated through three case studies. The system has passed its pilot study and is soon to be deployed in industry.

Hongming Zhang, Jiaxin Bai, Yan Song, Kun Xu, Changlong Yu, Yangqiu Song, Wilfred Ng, Dong Yu. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.