Artificial intelligence (AI) agents are systems capable of perceiving their environment, autonomously planning and executing tasks. Recent advancements in Large Language Models (LLMs) have introduced a transformative paradigm for AI agents, enabling them to interact with external resources and tools through prompt techniques. This advancement has significantly extended the capabilities of LLMs, positioning LLM-based AI Agents as an important research area. In such agents, the workflow integrates developer-written code, which manages framework construction and logic control, with LLM-generated natural language that enhances dynamic decision-making and interaction. However, inconsistencies between LLM outputs and developer logic can lead to defects, such as tool invocation failures. These issues introduce specific risks, leading to various defects in LLM-based AI Agents, including service interruptions and incorrect output. Despite the importance of these issues, there is a lack of systematic work that focuses on analyzing LLM-based AI Agents to uncover defects in their code. To address this gap, we present the first study focused on identifying and detecting defects in LLM Agents. We collected and analyzed 14,754 relevant developer reports from StackOverflow and GitHub. We further filtered 2,604 valid posts to define and classify eight types of agent code defects. Then, we designed a static analysis tool, named Agentable, to detect these defects. Agentable leverages Code Property Graphs (CPGs) and LLMs to analyze Agent workflows by efficiently identifying specific code patterns and analyzing natural language descriptions. To evaluate Agentable, we constructed two datasets: AgentSet, which consists of 84 real world Agent projects, and AgentTest, which contains 78 Agent projects specifically designed to include various types of defects. Our evaluation shows that Agentable achieves a precision of 88.79% on the real-world agent dataset and a recall of 91.03% on the manually labeled defect dataset. Furthermore, our analysis identifies 889 defects in real-world agent projects, highlighting the prevalence of these issues in practice.

Decentralized Finance (DeFi) staking is one of the most prominent applications within the DeFi ecosystem, where DeFi projects enable users to stake tokens on the platform and reward participants with additional tokens. However, logical defects in DeFi staking could enable attackers to claim unwarranted rewards by manipulating reward amounts, repeatedly claiming rewards, or engaging in other malicious actions. To mitigate these threats, we conducted the first study focused on defining and detecting logical defects in DeFi staking. Through the analysis of 64 security incidents and 144 audit reports, we identified six distinct types of logical defects, each accompanied by detailed descriptions and code examples. Building on this empirical research, we developed SSR (Safeguarding Staking Reward), a static analysis tool designed to detect logical defects in DeFi staking contracts. SSR utilizes a large language model (LLM) to extract fundamental information about staking logic and constructs a DeFi staking model. It then identifies logical defects by analyzing the model and the associated semantic features. We constructed a ground truth dataset based on known security incidents and audit reports to evaluate the effectiveness of SSR. The results indicate that SSR achieves an overall precision of 92.31%, a recall of 87.92%, and an F1-score of 88.85%. Additionally, to assess the prevalence of logical defects in real-world smart contracts, we compiled a large-scale dataset of 15,992 DeFi staking contracts. SSR detected that 3,557 (22.24%) of these contracts contained at least one logical defect.

Modeling interactive driving behaviors in complex scenarios remains a fundamental challenge for autonomous driving planning. Learning-based approaches attempt to address this challenge with advanced generative models, removing the dependency on over-engineered architectures for representation fusion. However, brute-force implementation by simply stacking transformer blocks lacks a dedicated mechanism for modeling interactive behaviors that is common in real driving scenarios. The scarcity of interactive driving data further exacerbates this problem, leaving conventional imitation learning methods ill-equipped to capture high-value interactive behaviors. We propose Flow Planner, which tackles these problems through coordinated innovations in data modeling, model architecture, and learning scheme. Specifically, we first introduce fine-grained trajectory tokenization, which decomposes the trajectory into overlapping segments to decrease the complexity of whole trajectory modeling. With a sophisticatedly designed architecture, we achieve efficient temporal and spatial fusion of planning and scene information, to better capture interactive behaviors. In addition, the framework incorporates flow matching with classifier-free guidance for multi-modal behavior generation, which dynamically reweights agent interactions during inference to maintain coherent response strategies, providing a critical boost for interactive scenario understanding. Experimental results on the large-scale nuPlan dataset demonstrate that Flow Planner achieves state-of-the-art performance among learning-based approaches while effectively modeling interactive behaviors in complex driving scenarios.

Reentrancy is one of the most notorious vulnerabilities in smart contracts, resulting in significant digital asset losses. However, many previous works indicate that current Reentrancy detection tools suffer from high false positive rates. Even worse, recent years have witnessed the emergence of new Reentrancy attack patterns fueled by intricate and diverse vulnerability exploit mechanisms. Unfortunately, current tools face a significant limitation in their capacity to adapt and detect these evolving Reentrancy patterns. Consequently, ensuring precise and highly extensible Reentrancy vulnerability detection remains critical challenges for existing tools. To address this issue, we propose a tool named ReEP, designed to reduce the false positives for Reentrancy vulnerability detection. Additionally, ReEP can integrate multiple tools, expanding its capacity for vulnerability detection. It evaluates results from existing tools to verify vulnerability likelihood and reduce false positives. ReEP also offers excellent extensibility, enabling the integration of different detection tools to enhance precision and cover different vulnerability attack patterns. We perform ReEP to eight existing state-of-the-art Reentrancy detection tools. The average precision of these eight tools increased from the original 0.5% to 73% without sacrificing recall. Furthermore, ReEP exhibits robust extensibility. By integrating multiple tools, the precision further improved to a maximum of 83.6%. These results demonstrate that ReEP effectively unites the strengths of existing works, enhances the precision of Reentrancy vulnerability detection tools.