Generative Engine Marketing (GEM) is an emerging ecosystem for monetizing generative engines, such as LLM-based chatbots, by seamlessly integrating relevant advertisements into their responses. At the core of GEM lies the generation and evaluation of ad-injected responses. However, existing benchmarks are not specifically designed for this purpose, which limits future research. To address this gap, we propose GEM-Bench, the first comprehensive benchmark for ad-injected response generation in GEM. GEM-Bench includes three curated datasets covering both chatbot and search scenarios, a metric ontology that captures multiple dimensions of user satisfaction and engagement, and several baseline solutions implemented within an extensible multi-agent framework. Our preliminary results indicate that, while simple prompt-based methods achieve reasonable engagement such as click-through rate, they often reduce user satisfaction. In contrast, approaches that insert ads based on pre-generated ad-free responses help mitigate this issue but introduce additional overhead. These findings highlight the need for future research on designing more effective and efficient solutions for generating ad-injected responses in GEM. The benchmark and all related resources are publicly available at https://gem-bench.org/.

This paper tackles \textbf{open-ended deep research (OEDR)}, a complex challenge where AI agents must synthesize vast web-scale information into insightful reports. Current approaches are plagued by dual-fold limitations: static research pipelines that decouple planning from evidence acquisition and monolithic generation paradigms that include redundant, irrelevant evidence, suffering from hallucination issues and low citation accuracy. To address these challenges, we introduce \textbf{WebWeaver}, a novel dual-agent framework that emulates the human research process. The planner operates in a dynamic cycle, iteratively interleaving evidence acquisition with outline optimization to produce a comprehensive, citation-grounded outline linking to a memory bank of evidence. The writer then executes a hierarchical retrieval and writing process, composing the report section by section. By performing targeted retrieval of only the necessary evidence from the memory bank via citations for each part, it effectively mitigates long-context issues and citation hallucinations. Our framework establishes a new state-of-the-art across major OEDR benchmarks, including DeepResearch Bench, DeepConsult, and DeepResearchGym. These results validate our human-centric, iterative methodology, demonstrating that adaptive planning and focused synthesis are crucial for producing comprehensive, trusted, and well-structured reports.

Prompt engineering is crucial for fully leveraging large language models (LLMs), yet most existing optimization methods follow a single trajectory, resulting in limited adaptability, gradient conflicts, and high computational overhead. We propose MAPGD (Multi-Agent Prompt Gradient Descent), a novel framework that reconceptualizes prompt optimization as a collaborative process among specialized agents. Each agent focuses on a distinct refinement dimension, such as instruction clarity, example selection, format structure, or stylistic adaptation, and their contributions are coordinated through semantic gradient embedding, conflict detection, and fusion. To further enhance robustness and stability, MAPGD introduces two new mechanisms: Hypersphere Constrained Gradient Clustering (HCGC), which enforces angular margin constraints for compact and well-separated clusters, and Channel Adaptive Agent Weighting (CAAW), which dynamically reweights agent contributions based on validation performance. Experiments on classification and reasoning benchmarks show that MAPGD consistently surpasses single-agent and random baselines in both accuracy and efficiency. Ablation studies confirm the effectiveness of gradient fusion, agent specialization, and conflict resolution. Together, these components establish MAPGD as a unified, gradient-based, and interpretable framework for robust prompt optimization with theoretical convergence guarantees.

Ensuring the safe use of agentic systems requires a thorough understanding of the range of malicious behaviors these systems may exhibit when under attack. In this paper, we evaluate the robustness of LLM-based agentic systems against attacks that aim to elicit harmful actions from agents. To this end, we propose a novel taxonomy of harms for agentic systems and a novel benchmark, BAD-ACTS, for studying the security of agentic systems with respect to a wide range of harmful actions. BAD-ACTS consists of 4 implementations of agentic systems in distinct application environments, as well as a dataset of 188 high-quality examples of harmful actions. This enables a comprehensive study of the robustness of agentic systems across a wide range of categories of harmful behaviors, available tools, and inter-agent communication structures. Using this benchmark, we analyze the robustness of agentic systems against an attacker that controls one of the agents in the system and aims to manipulate other agents to execute a harmful target action. Our results show that the attack has a high success rate, demonstrating that even a single adversarial agent within the system can have a significant impact on the security. This attack remains effective even when agents use a simple prompting-based defense strategy. However, we additionally propose a more effective defense based on message monitoring. We believe that this benchmark provides a diverse testbed for the security research of agentic systems. The benchmark can be found at github.com/JNoether/BAD-ACTS

Current artificial intelligence systems show near-human-level capabilities when deployed in isolation. Systems of a few collaborating intelligent agents are being engineered to perform tasks collectively. This raises the question of whether robotic matter, where many learning and intelligent agents interact, shows emergence of collective behaviour. And if so, which kind of phenomena would such systems exhibit? Here, we study a paradigmatic model for robotic matter: a stochastic many-particle system in which each particle is endowed with a deep neural network that predicts its transitions based on the particles' environments. For a one-dimensional model, we show that robotic matter exhibits complex emergent phenomena, including transitions between long-lived learning regimes, the emergence of particle species, and frustration. We also find a density-dependent phase transition with signatures of criticality. Using active matter theory, we show that this phase transition is a consequence of self-organisation mediated by emergent inter-particle interactions. Our simple model captures key features of more complex forms of robotic systems.

Non-convex constrained optimizations are ubiquitous in robotic applications such as multi-agent navigation, UAV trajectory optimization, and soft robot simulation. For this problem class, conventional optimizers suffer from small step sizes and slow convergence. We propose BC-ADMM, a variant of Alternating Direction Method of Multiplier (ADMM), that can solve a class of non-convex constrained optimizations with biconvex constraint relaxation. Our algorithm allows larger step sizes by breaking the problem into small-scale sub-problems that can be easily solved in parallel. We show that our method has both theoretical convergence speed guarantees and practical convergence guarantees in the asymptotic sense. Through numerical experiments in a row of four robotic applications, we show that BC-ADMM has faster convergence than conventional gradient descent and Newton's method in terms of wall clock time.

Large Language Models (LLMs) as autonomous agents are increasingly tasked with solving complex, long-horizon problems. Aligning these agents via preference-based offline methods like Direct Preference Optimization (DPO) is a promising direction, yet it faces a critical granularity mismatch. Trajectory-level DPO provides a signal that is too coarse for precise credit assignment, while step-level DPO is often too myopic to capture the value of multi-step behaviors. To resolve this challenge, we introduce Hierarchical Preference Learning (HPL), a hierarchical framework that optimizes LLM agents by leveraging preference signals at multiple, synergistic granularities. While HPL incorporates trajectory- and step-level DPO for global and local policy stability, its core innovation lies in group-level preference optimization guided by a dual-layer curriculum. Our approach first decomposes expert trajectories into semantically coherent action groups and then generates contrasting suboptimal groups to enable preference learning at a fine-grained, sub-task level. Then, instead of treating all preference pairs equally, HPL introduces a curriculum scheduler that organizes the learning process from simple to complex. This curriculum is structured along two axes: the group length, representing sub-task complexity, and the sample difficulty, defined by the reward gap between preferred and dispreferred action groups. Experiments on three challenging agent benchmarks show that HPL outperforms existing state-of-the-art methods. Our analyses demonstrate that the hierarchical DPO loss effectively integrates preference signals across multiple granularities, while the dual-layer curriculum is crucial for enabling the agent to solve a wide range of tasks, from simple behaviors to complex multi-step sequences.

The growing electricity demand and increased use of smart appliances are placing new pressures on power grids, making efficient energy management more important than ever. The existing energy management systems often prioritize system efficiency (balanced energy demand and supply) at the expense of user comfort. This paper addresses this gap by proposing a novel decentralized multi-agent coordination-based demand-side management system. The proposed system enables individual agents to coordinate for demand-side energy optimization while improving the user comfort and maintaining the system efficiency. A key innovation of this work is the introduction of a slot exchange mechanism, where agents first receive optimized appliance-level energy consumption schedules and then coordinate with each other to adjust these schedules through slot exchanges. This approach improves user comfort even when agents show non-altruistic behaviour, and it scales well with large populations. The system also promotes fairness by balancing satisfaction levels across users. For performance evaluation, a real-world dataset is used, and the results demonstrate that the proposed slot exchange mechanism increases user comfort and fairness without raising system inefficiency cost, making it a practical and scalable solution for future smart grids.

Recent advances in large language models (LLMs) opened up new directions for leveraging the collective expertise of multiple LLMs. These methods, such as Mixture-of-Agents, typically employ additional inference steps to generate intermediate outputs, which are then used to produce the final response. While multi-agent inference can enhance response quality, it can significantly increase the time to first token (TTFT), posing a challenge for latency-sensitive applications and hurting user experience. To address this issue, we propose staircase streaming for low-latency multi-agent inference. Instead of waiting for the complete intermediate outputs from previous steps, we begin generating the final response as soon as we receive partial outputs from these steps. Experimental results demonstrate that staircase streaming reduces TTFT by up to 93% while maintaining response quality.

Test-time reasoning significantly enhances pre-trained AI agents' performance. However, it requires an explicit environment model, often unavailable or overly complex in real-world scenarios. While MuZero enables effective model learning for search in perfect information games, extending this paradigm to imperfect information games presents substantial challenges due to more nuanced look-ahead reasoning techniques and large number of states relevant for individual decisions. This paper introduces an algorithm LAMIR that learns an abstracted model of an imperfect information game directly from the agent-environment interaction. During test time, this trained model is used to perform look-ahead reasoning. The learned abstraction limits the size of each subgame to a manageable size, making theoretically principled look-ahead reasoning tractable even in games where previous methods could not scale. We empirically demonstrate that with sufficient capacity, LAMIR learns the exact underlying game structure, and with limited capacity, it still learns a valuable abstraction, which improves game playing performance of the pre-trained agents even in large games.