Xi Zhang

Shuyi Liu, Yuming Shang, Xi Zhang

Retrieval-Augmented Generation (RAG) has emerged as a powerful framework for enhancing the capabilities of Large Language Models (LLMs) by integrating retrieval-based methods with generative models. As external knowledge repositories continue to expand and the parametric knowledge within models becomes outdated, a critical challenge for RAG systems is resolving conflicts between retrieved external information and LLMs' internal knowledge, which can significantly compromise the accuracy and reliability of generated content. However, existing approaches to conflict resolution typically operate at the token or semantic level, often leading to fragmented and partial understanding of factual discrepancies between LLMs' knowledge and context, particularly in knowledge-intensive tasks. To address this limitation, we propose TruthfulRAG, the first framework that leverages Knowledge Graphs (KGs) to resolve factual-level knowledge conflicts in RAG systems. Specifically, TruthfulRAG constructs KGs by systematically extracting triples from retrieved content, utilizes query-based graph retrieval to identify relevant knowledge, and employs entropy-based filtering mechanisms to precisely locate conflicting elements and mitigate factual inconsistencies, thereby enabling LLMs to generate faithful and accurate responses. Extensive experiments reveal that TruthfulRAG outperforms existing methods, effectively alleviating knowledge conflicts and improving the robustness and trustworthiness of RAG systems.

Jinyuan Fang, Yanwen Peng, Xi Zhang et al. · cambridge

Recent advances in large language models have sparked growing interest in AI agents capable of solving complex, real-world tasks. However, most existing agent systems rely on manually crafted configurations that remain static after deployment, limiting their ability to adapt to dynamic and evolving environments. To this end, recent research has explored agent evolution techniques that aim to automatically enhance agent systems based on interaction data and environmental feedback. This emerging direction lays the foundation for self-evolving AI agents, which bridge the static capabilities of foundation models with the continuous adaptability required by lifelong agentic systems. In this survey, we provide a comprehensive review of existing techniques for self-evolving agentic systems. Specifically, we first introduce a unified conceptual framework that abstracts the feedback loop underlying the design of self-evolving agentic systems. The framework highlights four key components: System Inputs, Agent System, Environment, and Optimisers, serving as a foundation for understanding and comparing different strategies. Based on this framework, we systematically review a wide range of self-evolving techniques that target different components of the agent system. We also investigate domain-specific evolution strategies developed for specialised fields such as biomedicine, programming, and finance, where optimisation objectives are tightly coupled with domain constraints. In addition, we provide a dedicated discussion on the evaluation, safety, and ethical considerations for self-evolving agentic systems, which are critical to ensuring their effectiveness and reliability. This survey aims to provide researchers and practitioners with a systematic understanding of self-evolving AI agents, laying the foundation for the development of more adaptive, autonomous, and lifelong agentic systems.

Yuefei Lyu, Xiaoyu Yang, Jiaxin Liu et al.

Social networks are frequently polluted by rumors, which can be detected by advanced models such as graph neural networks. However, the models are vulnerable to attacks and understanding the vulnerabilities is critical to rumor detection in practice. To discover subtle vulnerabilities, we design a powerful attacking algorithm to camouflage rumors in social networks based on reinforcement learning that can interact with and attack any black-box detectors. The environment has exponentially large state spaces, high-order graph dependencies, and delayed noisy rewards, making the state-of-the-art end-to-end approaches difficult to learn features as large learning costs and expressive limitation of graph deep models. Instead, we design domain-specific features to avoid learning features and produce interpretable attack policies. To further speed up policy optimization, we devise: (i) a credit assignment method that decomposes delayed rewards to atomic attacking actions proportional to the their camouflage effects on target rumors; (ii) a time-dependent control variate to reduce reward variance due to large graphs and many attacking steps, supported by the reward variance analysis and a Bayesian analysis of the prediction distribution. On three real world datasets of rumor detection tasks, we demonstrate: (i) the effectiveness of the learned attacking policy compared to rule-based attacks and current end-to-end approaches; (ii) the usefulness of the proposed credit assignment strategy and variance reduction components; (iii) the interpretability of the policy when generating strong attacks via the case study.