A central challenge in explainable AI, particularly in the visual domain, is producing explanations grounded in human-understandable concepts. To tackle this, we introduce OCEAN (Object-Centric Explananda via Agent Negotiation), a novel, inherently interpretable framework built on object-centric representations and a transparent multi-agent reasoning process. The game-theoretic reasoning process drives agents to agree on coherent and discriminative evidence, resulting in a faithful and interpretable decision-making process. We train OCEAN end-to-end and benchmark it against standard visual classifiers and popular posthoc explanation tools like GradCAM and LIME across two diagnostic multi-object datasets. Our results demonstrate competitive performance with respect to state-of-the-art black-box models with a faithful reasoning process, which was reflected by our user study, where participants consistently rated OCEAN's explanations as more intuitive and trustworthy.

Game theory has grown into a major field over the past few decades, and poker has long served as one of its key case studies. Game-Theory-Optimal (GTO) provides strategies to avoid loss in poker, but pure GTO does not guarantee maximum profit. To this end, we aim to develop a model that outperforms GTO strategies to maximize profit in No Limit Holdem, in heads-up (two-player) and multi-way (more than two-player) situations. Our model finds the GTO foundation and goes further to exploit opponents. The model first navigates toward many simulated poker hands against itself and keeps adjusting its decisions until no action can reliably beat it, creating a strong baseline that is close to the theoretical best strategy. Then, it adapts by observing opponent behavior and adjusting its strategy to capture extra value accordingly. Our results indicate that Monte-Carlo Counterfactual Regret Minimization (CFR) performs best in heads-up situations and CFR remains the strongest method in most multi-way situations. By combining the defensive strength of GTO with real-time exploitation, our approach aims to show how poker agents can move from merely not losing to consistently winning against diverse opponents.

Autonomous agents for long-sequence Graphical User Interface tasks are hindered by sparse rewards and the intractable credit assignment problem. To address these challenges, we introduce GUI-Shepherd, a Process Reward Model that provides dense, step-by-step feedback to guide agents. GUI-Shepherd is trained on a diverse large-scale data set of $52$k interactions that features human-annotated scores and GPT-4o generated rationales, enabling it to serve both as a reward provider for RL training and as a verifier for inference. As far as we know, we are the first to conduct a systematic study of process supervision in GUI agents, across diverse settings from online long-horizon tasks to offline single-step prediction. On the online AndroidWorld benchmark, GUI-Shepherd improves success rate by $7.7$ points via multi-turn online PPO, significantly outperforming Outcome Reward Model based competitors. When used as an inference verifier, it brings $5.1$ points improvements. The benefits generalize to the offline AndroidControl benchmark, with gains of $2.2$ points as a reward provider and $4.3$ points as a verifier. Collectively, our results establish that high-fidelity process supervision is critical for building more capable GUI agents and present a generalizable solution.

Agentic systems consisting of multiple LLM-driven agents coordinating through tools and structured interactions, are increasingly deployed for complex reasoning and problem-solving tasks. At the same time, emerging low-code and template-based agent development platforms (e.g., Dify) enable users to rapidly build and orchestrate agentic systems, which we refer to as platform-orchestrated agentic systems. However, these systems are also fragile and it remains unclear how to systematically identify their potential failure root cause. This paper presents a study of root cause identification of these platform-orchestrated agentic systems. To support this initiative, we construct a dataset AgentFail containing 307 failure logs from ten agentic systems, each with fine-grained annotations linking failures to their root causes. We additionally utilize counterfactual reasoning-based repair strategy to ensure the reliability of the annotation. Building on the dataset, we develop a taxonomy that characterizes failure root causes and analyze their distribution across different platforms and task domains. Furthermore, we introduce a benchmark that leverages LLMs for automatically identifying root causes, in which we also utilize the proposed taxonomy as guidance for LLMs. Results show that the taxonomy can largely improve the performance, thereby confirming its utility. Nevertheless, the accuracy of root cause identification reaches at most 33.6%, which indicates that this task still remains challenging. In light of these results, we also provide actionable guidelines for building such agentic systems. In summary, this paper provides a reliable dataset of failure root cause for platform-orchestrated agentic systems, corresponding taxonomy and benchmark, which serves as a foundation for advancing the development of more reliable agentic systems.

Large language models (LLMs) have recently advanced in reasoning when optimized with reinforcement learning (RL) under verifiable rewards. Existing methods primarily rely on outcome-based supervision to strengthen internal LLM reasoning, often leading to inefficient exploration and sparse rewards. To mitigate this issue, we propose Expert-Assisted Policy Optimization (EAPO), a novel RL framework that enhances exploration by incorporating multi-turn interactions with external experts during training. Unlike prior methods, where policies reason in isolation, EAPO incentivizes the policy to adaptively determine when and how to consult experts, yielding richer reward signals and more reliable reasoning trajectories. External assistance ultimately internalizes expert knowledge into the policy model, amplifying the model's inherent reasoning capabilities. During evaluation, the policy model has been well-optimized to solve questions independently, producing improved reasoning paths and more accurate solutions. Experiments on mathematical reasoning benchmarks, including AIME 2024, AIME 2025, and AIMO 2025, show that EAPO consistently outperforms expert-assisted workflow, expert-distilled models, and RL baselines, with an average gain of 5 points over self-exploratory models.

Embodied agents can identify and report safety hazards in the home environments. Accurately evaluating their capabilities in home safety inspection tasks is curcial, but existing benchmarks suffer from two key limitations. First, they oversimplify safety inspection tasks by using textual descriptions of the environment instead of direct visual information, which hinders the accurate evaluation of embodied agents based on Vision-Language Models (VLMs). Second, they use a single, static viewpoint for environmental observation, which restricts the agents' free exploration and cause the omission of certain safety hazards, especially those that are occluded from a fixed viewpoint. To alleviate these issues, we propose HomeSafeBench, a benchmark with 12,900 data points covering five common home safety hazards: fire, electric shock, falling object, trips, and child safety. HomeSafeBench provides dynamic first-person perspective images from simulated home environments, enabling the evaluation of VLM capabilities for home safety inspection. By allowing the embodied agents to freely explore the room, HomeSafeBench provides multiple dynamic perspectives in complex environments for a more thorough inspection. Our comprehensive evaluation of mainstream VLMs on HomeSafeBench reveals that even the best-performing model achieves an F1-score of only 10.23%, demonstrating significant limitations in current VLMs. The models particularly struggle with identifying safety hazards and selecting effective exploration strategies. We hope HomeSafeBench will provide valuable reference and support for future research related to home security inspections. Our dataset and code will be publicly available soon.

The key to achieving autonomous driving lies in topology-aware perception, the structured understanding of the driving environment with an emphasis on lane topology and road semantics. This survey systematically reviews four core research directions under this theme: vectorized map construction, topological structure modeling, prior knowledge fusion, and language model-based perception. Across these directions, we observe a unifying trend: a paradigm shift from static, pre-built maps to dynamic, sensor-driven perception. Specifically, traditional static maps have provided semantic context for autonomous systems. However, they are costly to construct, difficult to update in real time, and lack generalization across regions, limiting their scalability. In contrast, dynamic representations leverage on-board sensor data for real-time map construction and topology reasoning. Each of the four research directions contributes to this shift through compact spatial modeling, semantic relational reasoning, robust domain knowledge integration, and multimodal scene understanding powered by pre-trained language models. Together, they pave the way for more adaptive, scalable, and explainable autonomous driving systems.

Effective guardrails are essential for safely deploying LLM-based agents in critical applications. Despite recent advances, existing guardrails suffer from two fundamental limitations: (i) they apply uniform guardrail policies to all users, ignoring that the same agent behavior can harm some users while being safe for others; (ii) they check each response in isolation, missing how risks evolve and accumulate across multiple interactions. To solve these issues, we propose PSG-Agent, a personalized and dynamic system for LLM-based agents. First, PSG-Agent creates personalized guardrails by mining the interaction history for stable traits and capturing real-time states from current queries, generating user-specific risk thresholds and protection strategies. Second, PSG-Agent implements continuous monitoring across the agent pipeline with specialized guards, including Plan Monitor, Tool Firewall, Response Guard, Memory Guardian, that track cross-turn risk accumulation and issue verifiable verdicts. Finally, we validate PSG-Agent in multiple scenarios including healthcare, finance, and daily life automation scenarios with diverse user profiles. It significantly outperform existing agent guardrails including LlamaGuard3 and AGrail, providing an executable and auditable path toward personalized safety for LLM-based agents.

Multi-turn agent systems based on Large Language Models (LLMs) have been increasingly popular for software engineering tasks. While LLM agents show decent effectiveness, the high computational cost of input tokens due to the ever-growing trajectory remains an efficiency concern for their applications. Efficiency is largely neglected in existing studies and agent products, and this paper fills the gap by introducing an inference-time trajectory reduction approach to reduce the cost of agents. Through analyzing existing agent trajectories, we demonstrate that useless, redundant, and expired information is widespread in all trajectories, which can be identified and reduced without harming the agent's performance. We then design a simple yet effective trajectory reduction approach, AgentDiet, which automatically removes such waste information. We implement AgentDiet on a top-performing coding agent, and the evaluation on two LLMs and two benchmarks shows that AgentDiet can reduce input tokens by 39.9% ~ 59.7%, or the final computational cost by 21.1% ~ 35.9%, while maintaining the same agent performance. This indicates that trajectory reduction is a promising direction for agent systems.

This paper presents SIMPOL (Simplified Policy Iteration), a modular numerical framework for solving continuous-time heterogeneous agent models. The core economic problem, the optimization of consumption and savings under idiosyncratic uncertainty, is formulated as a coupled system of partial differential equations: a Hamilton-Jacobi-Bellman (HJB) equation for the agent's optimal policy and a Fokker-Planck-Kolmogorov (FPK) equation for the stationary wealth distribution. SIMPOL addresses this system using Howard's policy iteration with an *upwind* finite difference scheme that guarantees stability. A distinctive contribution is a novel consumption policy post-processing module that imposes regularity through smoothing and a projection onto an economically plausible slope band, improving convergence and model behavior. The robustness and accuracy of SIMPOL are validated through a set of integrated diagnostics, including verification of contraction in the Wasserstein-2 metric and comparison with the analytical solution of the Merton model in the no-volatility case. The framework is shown to be not only computationally efficient but also to produce solutions consistent with economic and mathematical theory, offering a reliable tool for research in quantitative macroeconomics.