Graph2Eval: Automatic Multimodal Task Generation for Agents via Knowledge Graphs

As multimodal LLM-driven agents continue to advance in autonomy and generalization, evaluation based on static datasets can no longer adequately assess their true capabilities in dynamic environments and diverse tasks. Existing LLM-based synthetic data methods are largely designed for LLM training and evaluation, and thus cannot be directly applied to agent tasks that require tool use and interactive capabilities. While recent studies have explored automatic agent task generation with LLMs, most efforts remain limited to text or image analysis, without systematically modeling multi-step interactions in web environments. To address these challenges, we propose Graph2Eval, a knowledge graph-based framework that automatically generates both multimodal document comprehension tasks and web interaction tasks, enabling comprehensive evaluation of agents' reasoning, collaboration, and interactive capabilities. In our approach, knowledge graphs constructed from multi-source external data serve as the task space, where we translate semantic relations into structured multimodal tasks using subgraph sampling, task templates, and meta-paths. A multi-stage filtering pipeline based on node reachability, LLM scoring, and similarity analysis is applied to guarantee the quality and executability of the generated tasks. Furthermore, Graph2Eval supports end-to-end evaluation of multiple agent types (Single-Agent, Multi-Agent, Web Agent) and measures reasoning, collaboration, and interaction capabilities. We instantiate the framework with Graph2Eval-Bench, a curated dataset of 1,319 tasks spanning document comprehension and web interaction scenarios. Experiments show that Graph2Eval efficiently generates tasks that differentiate agent and model performance, revealing gaps in reasoning, collaboration, and web interaction across different settings and offering a new perspective for agent evaluation.

OS Agents: A Survey on MLLM-based Agents for General Computing Devices Use

The dream to create AI assistants as capable and versatile as the fictional J.A.R.V.I.S from Iron Man has long captivated imaginations. With the evolution of (multi-modal) large language models ((M)LLMs), this dream is closer to reality, as (M)LLM-based Agents using computing devices (e.g., computers and mobile phones) by operating within the environments and interfaces (e.g., Graphical User Interface (GUI)) provided by operating systems (OS) to automate tasks have significantly advanced. This paper presents a comprehensive survey of these advanced agents, designated as OS Agents. We begin by elucidating the fundamentals of OS Agents, exploring their key components including the environment, observation space, and action space, and outlining essential capabilities such as understanding, planning, and grounding. We then examine methodologies for constructing OS Agents, focusing on domain-specific foundation models and agent frameworks. A detailed review of evaluation protocols and benchmarks highlights how OS Agents are assessed across diverse tasks. Finally, we discuss current challenges and identify promising directions for future research, including safety and privacy, personalization and self-evolution. This survey aims to consolidate the state of OS Agents research, providing insights to guide both academic inquiry and industrial development. An open-source GitHub repository is maintained as a dynamic resource to foster further innovation in this field. We present a 9-page version of our work, accepted by ACL 2025, to provide a concise overview to the domain.

HarmonyGuard: Toward Safety and Utility in Web Agents via Adaptive Policy Enhancement and Dual-Objective Optimization

Large language models enable agents to autonomously perform tasks in open web environments. However, as hidden threats within the web evolve, web agents face the challenge of balancing task performance with emerging risks during long-sequence operations. Although this challenge is critical, current research remains limited to single-objective optimization or single-turn scenarios, lacking the capability for collaborative optimization of both safety and utility in web environments. To address this gap, we propose HarmonyGuard, a multi-agent collaborative framework that leverages policy enhancement and objective optimization to jointly improve both utility and safety. HarmonyGuard features a multi-agent architecture characterized by two fundamental capabilities: (1) Adaptive Policy Enhancement: We introduce the Policy Agent within HarmonyGuard, which automatically extracts and maintains structured security policies from unstructured external documents, while continuously updating policies in response to evolving threats. (2) Dual-Objective Optimization: Based on the dual objectives of safety and utility, the Utility Agent integrated within HarmonyGuard performs the Markovian real-time reasoning to evaluate the objectives and utilizes metacognitive capabilities for their optimization. Extensive evaluations on multiple benchmarks show that HarmonyGuard improves policy compliance by up to 38% and task completion by up to 20% over existing baselines, while achieving over 90% policy compliance across all tasks. Our project is available here: https://github.com/YurunChen/HarmonyGuard.

Multimodal LLM-Guided Semantic Correction in Text-to-Image Diffusion

Diffusion models have become the mainstream architecture for text-to-image generation, achieving remarkable progress in visual quality and prompt controllability. However, current inference pipelines generally lack interpretable semantic supervision and correction mechanisms throughout the denoising process. Most existing approaches rely solely on post-hoc scoring of the final image, prompt filtering, or heuristic resampling strategies-making them ineffective in providing actionable guidance for correcting the generative trajectory. As a result, models often suffer from object confusion, spatial errors, inaccurate counts, and missing semantic elements, severely compromising prompt-image alignment and image quality. To tackle these challenges, we propose MLLM Semantic-Corrected Ping-Pong-Ahead Diffusion (PPAD), a novel framework that, for the first time, introduces a Multimodal Large Language Model (MLLM) as a semantic observer during inference. PPAD performs real-time analysis on intermediate generations, identifies latent semantic inconsistencies, and translates feedback into controllable signals that actively guide the remaining denoising steps. The framework supports both inference-only and training-enhanced settings, and performs semantic correction at only extremely few diffusion steps, offering strong generality and scalability. Extensive experiments demonstrate PPAD's significant improvements.

AppealCase: A Dataset and Benchmark for Civil Case Appeal Scenarios

Recent advances in LegalAI have primarily focused on individual case judgment analysis, often overlooking the critical appellate process within the judicial system. Appeals serve as a core mechanism for error correction and ensuring fair trials, making them highly significant both in practice and in research. To address this gap, we present the AppealCase dataset, consisting of 10,000 pairs of real-world, matched first-instance and second-instance documents across 91 categories of civil cases. The dataset also includes detailed annotations along five dimensions central to appellate review: judgment reversals, reversal reasons, cited legal provisions, claim-level decisions, and whether there is new information in the second instance. Based on these annotations, we propose five novel LegalAI tasks and conduct a comprehensive evaluation across 20 mainstream models. Experimental results reveal that all current models achieve less than 50% F1 scores on the judgment reversal prediction task, highlighting the complexity and challenge of the appeal scenario. We hope that the AppealCase dataset will spur further research in LegalAI for appellate case analysis and contribute to improving consistency in judicial decision-making.

AEIA-MN: Evaluating the Robustness of Multimodal LLM-Powered Mobile Agents Against Active Environmental Injection Attacks

As researchers continuously optimize AI agents to perform tasks more effectively within operating systems, they often neglect to address the critical need for enabling these agents to identify "impostors" within the system. Through an analysis of the agents' operating environment, we identified a potential threat: attackers can disguise their attack methods as environmental elements, injecting active disturbances into the agents' execution process, thereby disrupting their decision-making. We define this type of attack as Active Environment Injection Attack (AEIA). Based on this, we propose AEIA-MN, an active environment injection attack scheme that exploits interaction vulnerabilities in the mobile operating system to evaluate the robustness of MLLM-based agents against such threats. Experimental results show that even advanced MLLMs are highly vulnerable to this attack, achieving a maximum attack success rate of 93% in the AndroidWorld benchmark.

Personalized Federated Learning with Adaptive Feature Aggregation and Knowledge Transfer

Federated Learning(FL) is popular as a privacy-preserving machine learning paradigm for generating a single model on decentralized data. However, statistical heterogeneity poses a significant challenge for FL. As a subfield of FL, personalized FL (pFL) has attracted attention for its ability to achieve personalized models that perform well on non-independent and identically distributed (Non-IID) data. However, existing pFL methods are limited in terms of leveraging the global model's knowledge to enhance generalization while achieving personalization on local data. To address this, we proposed a new method personalized Federated learning with Adaptive Feature Aggregation and Knowledge Transfer (FedAFK), to train better feature extractors while balancing generalization and personalization for participating clients, which improves the performance of personalized models on Non-IID data. We conduct extensive experiments on three datasets in two widely-used heterogeneous settings and show the superior performance of our proposed method over thirteen state-of-the-art baselines.

Unveiling Privacy Vulnerabilities: Investigating the Role of Structure in Graph Data

The public sharing of user information opens the door for adversaries to infer private data, leading to privacy breaches and facilitating malicious activities. While numerous studies have concentrated on privacy leakage via public user attributes, the threats associated with the exposure of user relationships, particularly through network structure, are often neglected. This study aims to fill this critical gap by advancing the understanding and protection against privacy risks emanating from network structure, moving beyond direct connections with neighbors to include the broader implications of indirect network structural patterns. To achieve this, we first investigate the problem of Graph Privacy Leakage via Structure (GPS), and introduce a novel measure, the Generalized Homophily Ratio, to quantify the various mechanisms contributing to privacy breach risks in GPS. Based on this insight, we develop a novel graph private attribute inference attack, which acts as a pivotal tool for evaluating the potential for privacy leakage through network structures under worst-case scenarios. To protect users' private data from such vulnerabilities, we propose a graph data publishing method incorporating a learnable graph sampling technique, effectively transforming the original graph into a privacy-preserving version. Extensive experiments demonstrate that our attack model poses a significant threat to user privacy, and our graph data publishing method successfully achieves the optimal privacy-utility trade-off compared to baselines.

Unsupervised Distance Metric Learning for Anomaly Detection Over Multivariate Time Series

Distance-based time series anomaly detection methods are prevalent due to their relative non-parametric nature and interpretability. However, the commonly used Euclidean distance is sensitive to noise. While existing works have explored dynamic time warping (DTW) for its robustness, they only support supervised tasks over multivariate time series (MTS), leaving a scarcity of unsupervised methods. In this work, we propose FCM-wDTW, an unsupervised distance metric learning method for anomaly detection over MTS, which encodes raw data into latent space and reveals normal dimension relationships through cluster centers. FCM-wDTW introduces locally weighted DTW into fuzzy C-means clustering and learns the optimal latent space efficiently, enabling anomaly identification via data reconstruction. Experiments with 11 different types of benchmarks demonstrate our method's competitive accuracy and efficiency.

SourceP: Smart Ponzi Schemes Detection on Ethereum Using Pre-training Model with Data Flow

As blockchain technology becomes more and more popular, a typical financial scam, the Ponzi scheme, has also emerged in the blockchain platform Ethereum. This Ponzi scheme deployed through smart contracts, also known as the smart Ponzi scheme, has caused a lot of economic losses and negative impacts. Existing methods for detecting smart Ponzi schemes on Ethereum mainly rely on bytecode features, opcode features, account features, and transaction behavior features of smart contracts, and such methods lack interpretability and sustainability. In this paper, we propose SourceP, a method to detect smart Ponzi schemes on the Ethereum platform using pre-training models and data flow, which only requires using the source code of smart contracts as features to explore the possibility of detecting smart Ponzi schemes from another direction. SourceP reduces the difficulty of data acquisition and feature extraction of existing detection methods while increasing the interpretability of the model. Specifically, we first convert the source code of a smart contract into a data flow graph and then introduce a pre-training model based on learning code representations to build a classification model to identify Ponzi schemes in smart contracts. The experimental results show that SourceP achieves 87.2\% recall and 90.7\% F-score for detecting smart Ponzi schemes within Ethereum's smart contract dataset, outperforming state-of-the-art methods in terms of performance and sustainability. We also demonstrate through additional experiments that pre-training models and data flow play an important contribution to SourceP, as well as proving that SourceP has a good generalization ability.