Recent advancements in LLM-based multi-agent (LLM-MA) systems have shown promise, yet significant challenges remain in managing communication and refinement when agents collaborate on complex tasks. In this paper, we propose \textit{Talk Structurally, Act Hierarchically (TalkHier)}, a novel framework that introduces a structured communication protocol for context-rich exchanges and a hierarchical refinement system to address issues such as incorrect outputs, falsehoods, and biases. \textit{TalkHier} surpasses various types of SoTA, including inference scaling model (OpenAI-o1), open-source multi-agent models (e.g., AgentVerse), and majority voting strategies on current LLM and single-agent baselines (e.g., ReAct, GPT4o), across diverse tasks, including open-domain question answering, domain-specific selective questioning, and practical advertisement text generation. These results highlight its potential to set a new standard for LLM-MA systems, paving the way for more effective, adaptable, and collaborative multi-agent frameworks. The code is available https://github.com/sony/talkhier.

Effective budget allocation is crucial for optimizing the performance of digital advertising campaigns. However, the development of practical budget allocation algorithms remain limited, primarily due to the lack of public datasets and comprehensive simulation environments capable of verifying the intricacies of real-world advertising. While multi-armed bandit (MAB) algorithms have been extensively studied, their efficacy diminishes in non-stationary environments where quick adaptation to changing market dynamics is essential. In this paper, we advance the field of budget allocation in digital advertising by introducing three key contributions. First, we develop a simulation environment designed to mimic multichannel advertising campaigns over extended time horizons, incorporating logged real-world data. Second, we propose an enhanced combinatorial bandit budget allocation strategy that leverages a saturating mean function and a targeted exploration mechanism with change-point detection. This approach dynamically adapts to changing market conditions, improving allocation efficiency by filtering target regions based on domain knowledge. Finally, we present both theoretical analysis and empirical results, demonstrating that our method consistently outperforms baseline strategies, achieving higher rewards and lower regret across multiple real-world campaigns.

Online advertising has become one of the most successful business models of the internet era. Impression opportunities are typically allocated through real-time auctions, where advertisers bid to secure advertisement slots. Deciding the best bid for an impression opportunity is challenging, due to the stochastic nature of user behavior and the variability of advertisement traffic over time. In this work, we propose a framework for training auto-bidding agents in multi-slot second-price auctions to maximize acquisitions (e.g., clicks, conversions) while adhering to budget and cost-per-acquisition (CPA) constraints. We exploit the insight that, after an advertisement campaign concludes, determining the optimal bids for each impression opportunity can be framed as a multiple-choice knapsack problem (MCKP) with a nonlinear objective. We propose an "oracle" algorithm that identifies a near-optimal combination of impression opportunities and advertisement slots, considering both past and future advertisement traffic data. This oracle solution serves as a training target for a student network which bids having access only to real-time information, a method we term Oracle Imitation Learning (OIL). Through numerical experiments, we demonstrate that OIL achieves superior performance compared to both online and offline reinforcement learning algorithms, offering improved sample efficiency. Notably, OIL shifts the complexity of training auto-bidding agents from crafting sophisticated learning algorithms to solving a nonlinear constrained optimization problem efficiently.

Current keyword decision-making in sponsored search advertising relies on large, static datasets, limiting the ability to automatically set up keywords and adapt to real-time KPI metrics and product updates that are essential for effective advertising. In this paper, we propose On-the-fly Keyword Generation (OKG), an LLM agent-based method that dynamically monitors KPI changes and adapts keyword generation in real time, aligning with strategies recommended by advertising platforms. Additionally, we introduce the first publicly accessible dataset containing real keyword data along with its KPIs across diverse domains, providing a valuable resource for future research. Experimental results show that OKG significantly improves keyword adaptability and responsiveness compared to traditional methods. The code for OKG and the dataset are available at https://github.com/sony/okg.

With the ability to learn from static datasets, Offline Reinforcement Learning (RL) emerges as a compelling avenue for real-world applications. However, state-of-the-art offline RL algorithms perform sub-optimally when confronted with limited data confined to specific regions within the state space. The performance degradation is attributed to the inability of offline RL algorithms to learn appropriate actions for rare or unseen observations. This paper proposes a novel domain knowledge-based regularization technique and adaptively refines the initial domain knowledge to considerably boost performance in limited data with partially omitted states. The key insight is that the regularization term mitigates erroneous actions for sparse samples and unobserved states covered by domain knowledge. Empirical evaluations on standard discrete environment datasets demonstrate a substantial average performance increase of at least 27% compared to existing offline RL algorithms operating on limited data.

Recent advancements in offline Reinforcement Learning (Offline RL) have led to an increased focus on methods based on conservative policy updates to address the Out-of-Distribution (OOD) issue. These methods typically involve adding behavior regularization or modifying the critic learning objective, focusing primarily on states or actions with substantial dataset support. However, we challenge this prevailing notion by asserting that the absence of an action or state from a dataset does not necessarily imply its suboptimality. In this paper, we propose a novel approach to tackle the OOD problem. We introduce an offline RL teacher-student framework, complemented by a policy similarity measure. This framework enables the student policy to gain insights not only from the offline RL dataset but also from the knowledge transferred by a teacher policy. The teacher policy is trained using another dataset consisting of state-action pairs, which can be viewed as practical domain knowledge acquired without direct interaction with the environment. We believe this additional knowledge is key to effectively solving the OOD issue. This research represents a significant advancement in integrating a teacher-student network into the actor-critic framework, opening new avenues for studies on knowledge transfer in offline RL and effectively addressing the OOD challenge.