Weixing Shen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Weixing Shen has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
Huanghai Liu
Tsinghua University
4 shared publicationsYiran Hu
Tsinghua University University of Waterloo
4 shared publicationsQingjing Chen
Tsinghua University
3 shared publicationsYun Liu
Tsinghua University
3 shared publications- Charles L. A. Clarke
University of Waterloo
2 shared publications Chong Wang
Tsinghua University
2 shared publications
Latest Publications
- Open Access
- 11 April 2025
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (27) , 28106-28115
J&H: Evaluating the Robustness of Large Language Models Under Knowledge-Injection Attacks in Legal Domain
As the scale and capabilities of Large Language Models (LLMs) increase, their applications in knowledge-intensive fields such as legal domain have garnered widespread attention. However, it remains doubtful whether these LLMs make judgments based on domain knowledge for reasoning. If LLMs base their judgments solely on specific words or patterns, rather than on the underlying logic of the language, the “LLM-as-judges” paradigm poses substantial risks in the real-world applications. To address this question, we propose a method of legal knowledge injection attacks for robustness testing, thereby inferring whether LLMs have learned legal knowledge and reasoning logic. In this paper, we propose J&H: an evaluation framework for detecting the robustness of LLMs under knowledge injection attacks in the legal domain. The aim of the framework is to explore whether LLMs perform deductive reasoning when accomplishing legal tasks. To further this aim, we have attacked each part of the reasoning logic underlying these tasks (major premise, minor premise, and conclusion generation). We have collected mistakes that legal experts might make in judicial decisions in the real world, such as typos, legal synonyms, inaccurate external legal statutes retrieval. However, in real legal practice, legal experts tend to overlook these mistakes and make judgments based on logic. However, when faced with these errors, LLMs are likely to be misled by typographical errors and may not utilize logic in their judgments. We conducted knowledge injection attacks on existing general and domain-specific LLMs. Current LLMs are not robust against the attacks employed in our experiments. In addition we propose and compare several methods to enhance the knowledge robustness of LLMs. All code can be found at the link.
- 11 April 2025
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (27) , 28106-28115
Open AccessJ&H: Evaluating the Robustness of Large Language Models Under Knowledge-Injection Attacks in Legal Domain
As the scale and capabilities of Large Language Models (LLMs) increase, their applications in knowledge-intensive fields such as legal domain have garnered widespread attention. However, it remains doubtful whether these LLMs make judgments based on domain knowledge for reasoning. If LLMs base their judgments solely on specific words or patterns, rather than on the underlying logic of the language, the “LLM-as-judges” paradigm poses substantial risks in the real-world applications. To address this question, we propose a method of legal knowledge injection attacks for robustness testing, thereby inferring whether LLMs have learned legal knowledge and reasoning logic. In this paper, we propose J&H: an evaluation framework for detecting the robustness of LLMs under knowledge injection attacks in the legal domain. The aim of the framework is to explore whether LLMs perform deductive reasoning when accomplishing legal tasks. To further this aim, we have attacked each part of the reasoning logic underlying these tasks (major premise, minor premise, and conclusion generation). We have collected mistakes that legal experts might make in judicial decisions in the real world, such as typos, legal synonyms, inaccurate external legal statutes retrieval. However, in real legal practice, legal experts tend to overlook these mistakes and make judgments based on logic. However, when faced with these errors, LLMs are likely to be misled by typographical errors and may not utilize logic in their judgments. We conducted knowledge injection attacks on existing general and domain-specific LLMs. Current LLMs are not robust against the attacks employed in our experiments. In addition we propose and compare several methods to enhance the knowledge robustness of LLMs. All code can be found at the link.
- Open Access
J&H: Evaluating the Robustness of Large Language Models Under Knowledge-Injection Attacks in Legal Domain
As the scale and capabilities of Large Language Models (LLMs) increase, their applications in knowledge-intensive fields such as legal domain have garnered widespread attention. However, it remains doubtful whether these LLMs make judgments based on domain knowledge for reasoning. If LLMs base their judgments solely on specific words or patterns, rather than on the underlying logic of the language, the ''LLM-as-judges'' paradigm poses substantial risks in the real-world applications. To address this question, we propose a method of legal knowledge injection attacks for robustness testing, thereby inferring whether LLMs have learned legal knowledge and reasoning logic. In this paper, we propose J&H: an evaluation framework for detecting the robustness of LLMs under knowledge injection attacks in the legal domain. The aim of the framework is to explore whether LLMs perform deductive reasoning when accomplishing legal tasks. To further this aim, we have attacked each part of the reasoning logic underlying these tasks (major premise, minor premise, and conclusion generation). We have collected mistakes that legal experts might make in judicial decisions in the real world, such as typos, legal synonyms, inaccurate external legal statutes retrieval. However, in real legal practice, legal experts tend to overlook these mistakes and make judgments based on logic. However, when faced with these errors, LLMs are likely to be misled by typographical errors and may not utilize logic in their judgments. We conducted knowledge injection attacks on existing general and domain-specific LLMs. Current LLMs are not robust against the attacks employed in our experiments. In addition we propose and compare several methods to enhance the knowledge robustness of LLMs.
Open AccessJ&H: Evaluating the Robustness of Large Language Models Under Knowledge-Injection Attacks in Legal Domain
As the scale and capabilities of Large Language Models (LLMs) increase, their applications in knowledge-intensive fields such as legal domain have garnered widespread attention. However, it remains doubtful whether these LLMs make judgments based on domain knowledge for reasoning. If LLMs base their judgments solely on specific words or patterns, rather than on the underlying logic of the language, the ''LLM-as-judges'' paradigm poses substantial risks in the real-world applications. To address this question, we propose a method of legal knowledge injection attacks for robustness testing, thereby inferring whether LLMs have learned legal knowledge and reasoning logic. In this paper, we propose J&H: an evaluation framework for detecting the robustness of LLMs under knowledge injection attacks in the legal domain. The aim of the framework is to explore whether LLMs perform deductive reasoning when accomplishing legal tasks. To further this aim, we have attacked each part of the reasoning logic underlying these tasks (major premise, minor premise, and conclusion generation). We have collected mistakes that legal experts might make in judicial decisions in the real world, such as typos, legal synonyms, inaccurate external legal statutes retrieval. However, in real legal practice, legal experts tend to overlook these mistakes and make judgments based on logic. However, when faced with these errors, LLMs are likely to be misled by typographical errors and may not utilize logic in their judgments. We conducted knowledge injection attacks on existing general and domain-specific LLMs. Current LLMs are not robust against the attacks employed in our experiments. In addition we propose and compare several methods to enhance the knowledge robustness of LLMs.
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 3794-3814
JUREX-4E: Juridical Expert-Annotated Four-Element Knowledge Base for Legal Reasoning
Huanghai Liu, Quzhe Huang, Qingjing Chen, Yiran Hu, Jiayu Ma, Yun Liu, Weixing Shen, Yansong Feng. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 3794-3814
Open AccessJUREX-4E: Juridical Expert-Annotated Four-Element Knowledge Base for Legal Reasoning
Huanghai Liu, Quzhe Huang, Qingjing Chen, Yiran Hu, Jiayu Ma, Yun Liu, Weixing Shen, Yansong Feng. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.
- Open Access
- 1 August 2024
- Published by International Joint Conferences on Artificial Intelligence
- Vol. 8, 7491-7499
DASS: A Dual-Branch Attention-based Framework for Trajectory Similarity Learning with Spatial and Semantic Fusion
Trajectory similarity aims to identify pairs of similar trajectories, serving as a crucial operation in spatial-temporal data mining. Although several approaches have been proposed, they encounter the following two issues: 1) An overemphasis on spatial similarity in road networks while the rich semantic information embedded in trajectories is not fully exploited; 2) Dependence on Recurrent Neural Network (RNN) architectures would struggle to capture long-term dependencies. To address these limitations, we propose a Dual-branch Attention-based framework with Spatial and Semantic information (DASS) based on self-supervised learning. Specifically, DASS comprises two core components: 1) A trajectory representation module that models spatial-temporal adjacent relationships in the form of graph and converts semantics into numerical embeddings. 2) A backbone encoder with a co-attention module to independently process two features before they are integrated. Extensive experiments on real-world datasets demonstrate that DASS outperforms state-of-the-art methods, establishing itself as a novel paradigm.
- 1 August 2024
- Published by International Joint Conferences on Artificial Intelligence
- Vol. 8, 7491-7499
Open AccessDASS: A Dual-Branch Attention-based Framework for Trajectory Similarity Learning with Spatial and Semantic Fusion
Trajectory similarity aims to identify pairs of similar trajectories, serving as a crucial operation in spatial-temporal data mining. Although several approaches have been proposed, they encounter the following two issues: 1) An overemphasis on spatial similarity in road networks while the rich semantic information embedded in trajectories is not fully exploited; 2) Dependence on Recurrent Neural Network (RNN) architectures would struggle to capture long-term dependencies. To address these limitations, we propose a Dual-branch Attention-based framework with Spatial and Semantic information (DASS) based on self-supervised learning. Specifically, DASS comprises two core components: 1) A trajectory representation module that models spatial-temporal adjacent relationships in the form of graph and converts semantics into numerical embeddings. 2) A backbone encoder with a co-attention module to independently process two features before they are integrated. Extensive experiments on real-world datasets demonstrate that DASS outperforms state-of-the-art methods, establishing itself as a novel paradigm.
- 13 July 2017
- Published by Cambridge University Press (CUP) in Asian Journal of Comparative Law
- Vol. 12 (2) , 371-400
Guardianship and Health Decisions in China and Australia: A Comparative Analysis
This article compares the Australian and Chinese adult guardianship systems, and considers whether there is potential for drawing on some (or many) aspects of the Australian model for the Chinese legal framework. Australia has a well-developed guardianship framework that provides mechanisms for making healthcare decisions when an adult is no longer able to do so. This framework has evolved over many years and, in some cases, individuals can decide about medical treatment in advance of the situation arising, or who should be the decision-maker if he or she later loses capacity. The current Chinese legal framework, on the other hand, is a fragmented one and comprises laws that were not designed to deal with how healthcare decisions can be made for a person without capacity. This article outlines the legal framework in both jurisdictions and considers whether, having regard to the fact that these two countries have different values and cultures, there are features of the Australian guardianship system that could inform the development of Chinese law.
- 13 July 2017
- Published by Cambridge University Press (CUP) in Asian Journal of Comparative Law
- Vol. 12 (2) , 371-400
Guardianship and Health Decisions in China and Australia: A Comparative Analysis
This article compares the Australian and Chinese adult guardianship systems, and considers whether there is potential for drawing on some (or many) aspects of the Australian model for the Chinese legal framework. Australia has a well-developed guardianship framework that provides mechanisms for making healthcare decisions when an adult is no longer able to do so. This framework has evolved over many years and, in some cases, individuals can decide about medical treatment in advance of the situation arising, or who should be the decision-maker if he or she later loses capacity. The current Chinese legal framework, on the other hand, is a fragmented one and comprises laws that were not designed to deal with how healthcare decisions can be made for a person without capacity. This article outlines the legal framework in both jurisdictions and considers whether, having regard to the fact that these two countries have different values and cultures, there are features of the Australian guardianship system that could inform the development of Chinese law.