Although LLM-based agents, powered by Large Language Models (LLMs), can use external tools and memory mechanisms to solve complex real-world tasks, they may also introduce critical security vulnerabilities. However, the existing literature does not comprehensively evaluate attacks and defenses against LLM-based agents. To address this, we introduce Agent Security Bench (ASB), a comprehensive framework designed to formalize, benchmark, and evaluate the attacks and defenses of LLM-based agents, including 10 scenarios (e.g., e-commerce, autonomous driving, finance), 10 agents targeting the scenarios, over 400 tools, 27 different types of attack/defense methods, and 7 evaluation metrics. Based on ASB, we benchmark 10 prompt injection attacks, a memory poisoning attack, a novel Plan-of-Thought backdoor attack, 4 mixed attacks, and 11 corresponding defenses across 13 LLM backbones. Our benchmark results reveal critical vulnerabilities in different stages of agent operation, including system prompt, user prompt handling, tool usage, and memory retrieval, with the highest average attack success rate of 84.30\%, but limited effectiveness shown in current defenses, unveiling important works to be done in terms of agent security for the community. We also introduce a new metric to evaluate the agents' capability to balance utility and security. Our code can be found at https://github.com/agiresearch/ASB.

Training and deploying reinforcement learning (RL) policies for robots is a complex task, requiring careful design of reward functions, sim-to-real transfer, and performance evaluation across various robot configurations. These tasks traditionally demand significant human expertise and effort. To address these challenges, this paper introduces Anybipe, a novel, fully automated, end-to-end framework for training and deploying bipedal robots, leveraging large language models (LLMs) for reward function generation, while supervising model training, evaluation, and deployment. The framework integrates comprehensive quantitative metrics to assess policy performance, deployment effectiveness, and safety. Additionally, it allows users to incorporate prior knowledge and preferences, improving the accuracy and alignment of generated policies with expectations. We demonstrate how Anybipe reduces human labor while maintaining high levels of accuracy and safety, examined on three different bipedal robots, showcasing its potential for autonomous RL training and deployment.

Class-incremental fault diagnosis requires a model to adapt to new fault classes while retaining previous knowledge. However, limited research exists for imbalanced and long-tailed data. Extracting discriminative features from few-shot fault data is challenging, and adding new fault classes often demands costly model retraining. Moreover, incremental training of existing methods risks catastrophic forgetting, and severe class imbalance can bias the model's decisions toward normal classes. To tackle these issues, we introduce a supervised contrastive knowledge distillation for class incremental fault diagnosis (SCLIFD) framework proposing supervised contrastive knowledge distillation for improved representation learning capability and less forgetting, a novel prioritized exemplar selection method for sample replay to alleviate catastrophic forgetting, and the random forest classifier to address the class imbalance. Extensive experimentation on simulated and real-world industrial datasets across various imbalance ratios demonstrates the superiority of SCLIFD over existing approaches.

When data mining occurs on distributed data, privacy of parties becomes great concerns. This paper considers the problem of mining quantitative association rules without revealing the private information of parties who compute jointly and share distributed data. The issue is an area of privacy preserving data mining (PPDM) research. Some researchers have considered the case of mining Boolean association rules; however, this method cannot be easily applied to quantitative rules mining. A new secure set union algorithm is proposed in this paper, which unifies the input sets of parties without revealing any element's owner and has lower time cost than existing algorithms. The new algorithm takes the advantages of both in privacy-preserving Boolean association rules mining and in privacy-preserving quantitative association mining. This paper also presents an algorithm for privacy-preserving quantitative association rules mining over horizontally portioned data, based on CF tree and secure sum algorithm. Besides, the analysis of the correctness, the security and the complexity of our algorithms are provided