张瑶

张瑶，南开大学统计与数据科学学院，讲师。她先后于南开大学获得学士与博士学位，期间曾赴新加坡国立大学NExT研究中心进行访问学习。她的研究聚焦于自然语言处理、知识推理和对话系统等统计学与数据科学的前沿领域。目前，她主持一项国家自然科学基金项目，曾荣获天津市科技进步二等奖，并已在ACL、SIGIR、AAAI、EMNLP等国际顶级会议及期刊上发表了十余篇高水平论文。

课题组正在招收对科研充满热情的同学，欢迎将你的简历和想法发送至：yaozhang@nankai.edu.cn

个人主页：https://yaozhangnk.github.io/

2022年9月至今 南开大学统计与数据科学学院，讲师。

张瑶，南开大学统计与数据科学学院，讲师。她先后于南开大学获得学士与博士学位，期间曾赴新加坡国立大学 NExT 研究中心进行访问学习。她的研究聚焦于自然语言处理、知识推理和对话系统等统计学与数据科学的前沿领域。目前，她主持一项国家自然科学基金项目，曾荣获天津市科技进步二等奖，并已在 ACL、SIGIR、AAAI、EMNLP 等国际顶级会议及期刊上发表了十余篇高水平论文。

个人主页：https://yaozhangnk.github.io/

My research focuses on trustworthy knowledge reasoning that integrates LLMs with KGs, aiming to enhance the reasoning capability, reliability, and safety of models in complex knowledge-intensive scenarios. Specifically, my work covers the following directions:

Knowledge-Enhanced Reasoning

I study methods for integrating LLMs with KGs, leveraging structured knowledge to support complex multi-hop reasoning and improve the accuracy and interpretability of the reasoning process.

Reasoning Reliability

I investigate reasoning stability under conditions of dynamic knowledge updates, insufficient evidence, and knowledge conflicts. My work explores knowledge editing, conflict resolution, and proactive abstention mechanisms to reduce hallucinations and erroneous reasoning.

Reasoning Safety

I study safety risks in knowledge reasoning scenarios, including jailbreak attacks, knowledge-path manipulation, and reasoning-chain interference, and develop corresponding evaluation and defense mechanisms.

Real-World Applications

In addition, our team has a long-standing interest in the deployment of AI technologies in real-world applications, particularly in industrial intelligence and healthcare. We explore how trustworthy knowledge reasoning can support practical needs such as complex decision support, risk identification, and human-AI collaboration.

《自然语言处理与文本挖掘》，本科生，2026|2025|2024

《自然语言处理与文本挖掘》，研究生，2026|2025|2024|2023

• Listening to Patients: Detecting and Mitigating Patient Misreport in Medical Dialogue System. In ACL, 2025. (CCF A)

• Generating Questions, Answers, and Distractors for Videos: Exploring Semantic Uncertainty of Object Motions. In ACL, 2025. (CCF A)

• SCOP: Evaluating the Comprehension Process of Large Language Models from a Cognitive View. In ACL, 2025. (CCF A)

• Enhancing Cross-Lingual Dialogue Summarization through Interpretable Chain-of-Thought. In DASFAA, 2025. (CCF B)

• Can We Learn Question, Answer, and Distractors All from an Image? A New Task for Multiple-choice Visual Question Answering. In LREC-COLING, 2024, pp. 2852--2863. (CCF B)

• Exploring Union and Intersection of Visual Regions for Generating Questions, Answers, and Distractors. In EMNLP, 2024, pp.1479--1489. (CCF B)

• Well Begun Is Half Done: Generator-agnostic Knowledge Pre-selection for Knowledge-grounded Dialogue. In EMNLP, 2023. (CCF B)

• Fact-Tree Reasoning for N-ary Question Answering over Knowledge Graphs. In ACL, 2022. (CCF A)

• Interacting with Non-Cooperative User: A New Paradigm for Proactive Dialogue Policy. In SIGIR, 2022. (CCF A)

• Modeling Temporal-Modal Entity Graph for Procedural Multimodal Machine Comprehension. In ACL, 2022. (CCF A)

• Generalized Relation Learning with Semantic Correlation Awareness for Link Prediction. In AAAI, 2021, pp.4679-4687. (CCF A)

• GMH: A General Multi-hop Reasoning Model for KG Completion. In EMNLP, 2021, pp.3437-3446. (CCF B)

• TRFR: A ternary relation link prediction framework on Knowledge graphs. In Ad Hoc Networks, 2021, vol.113, pp.102402. (SCI II)

• Spatiotemporal-aware region recommendation with deep metric learning. In DASFAA, 2019, pp.491-494. (CCF B)