Cooperative broadcast is an efficient approach to improve Wi-Fi broadcast performance in crowded scenarios with densely deployed access points (APs). However, existing concurrent transmission MAC protocols cannot perfectly synchronize APs for the receiving user, and the superimposed channels at users vary over time due to multi-path effects with different carrier frequency offsets (CFOs) from the APs. Traditional channel estimation methods, which treat the superimposed channels as a whole and use a portion of the superimposed channels to derive the rest, are unsuitable. To solve the problem, we propose a fine-grained channel estimation approach that first estimates channel taps and CFOs of each AP, and then reconstructs the superimposed channels. We first study a benchmark channel estimation algorithm that utilizes a widely adopted compressed sensing (CS) technique. However, through analysis and simulations, we show that the CS-based algorithm suffers from high correlation problems in the constructed sensing matrix and the non-sparse channel problem in practice, leading to an estimation error floor at high SNRs. To solve these problems, we present a two-stage channel estimation algorithm. It first estimates the CFOs by identifying the most likely CFO combination matching the received signals, and then estimates the time-domain channel taps. Simulation and experimental results show that the two-stage channel estimation algorithm achieves much lower bit error rate (BER) and packet error rate (PER) than the traditional IEEE 802.11 approach, and the two-stage algorithm outperforms the CS-based algorithm, especially at high SNRs. The network-layer simulation results further demonstrate that, empowered by the proposed two-stage channel estimation algorithm, the cooperative broadcast scheme improves throughput by at least 1.4× (up to 46.8×) compared with the unicast-based broadcast schemes, and by approximately 0.6× to 0.8× compared with the simple uncooperative broadcast scheme.

How the recent progress of reasoning large language models (LLMs), especially the new open-source model DeepSeek-R1, can benefit financial services is an underexplored problem. While LLMs have ignited numerous applications within the financial sector, including financial news analysis and general customer interactions, DeepSeek-R1 further unlocks the advanced reasoning ability with multiple reinforcement learning-integrated training steps for more complex financial queries and provides distilled student models for resource-constrained scenarios. In this paper, we first introduce the technological preliminaries of DeepSeek-R1. Subsequently, we benchmark the performance of DeepSeek-R1 and its distilled students on two public financial question–answer (QA) datasets as a starting point for interdisciplinary research on financial artificial intelligence (AI). Then, we discuss the opportunities that DeepSeek-R1 offers to current financial services, its current limitations, and three future research directions. In conclusion, we argue for a proper approach to adopting reasoning LLMs for financial AI.

Alpha mining, which refers to the systematic discovery of data-driven signals predictive of future cross-sectional returns, is a central task in quantitative research. Recent progress in large language models (LLMs) has sparked interest in LLM-based alpha mining frameworks, which offer a promising middle ground between human-guided and fully automated alpha mining approaches and deliver both speed and semantic depth. This study presents a structured review of emerging LLM-based alpha mining systems from an agentic perspective, and analyzes the functional roles of LLMs, ranging from miners and evaluators to interactive assistants. Despite early progress, key challenges remain, including simplified performance evaluation, limited numerical understanding, lack of diversity and originality, weak exploration dynamics, temporal data leakage, and black-box risks and compliance challenges. Accordingly, we outline future directions, including improving reasoning alignment, expanding to new data modalities, rethinking evaluation protocols, and integrating LLMs into more general-purpose quantitative systems. Our analysis suggests that LLM is a scalable interface for amplifying both domain expertise and algorithmic rigor, as it amplifies domain expertise by transforming qualitative hypotheses into testable factors and enhances algorithmic rigor for rapid backtesting and semantic reasoning. The result is a complementary paradigm, where intuition, automation, and language-based reasoning converge to redefine the future of quantitative research.

Federated Foundation Models (FedFMs) represent a distributed learning paradigm that fuses general competences of foundation models as well as privacy-preserving capabilities of federated learning. This combination allows the large foundation models and the small local domain models at the remote clients to learn from each other in a teacher-student learning setting. This paper provides a comprehensive summary of the ten challenging problems inherent in FedFMs, encompassing foundational theory, utilization of private data, continual learning, unlearning, Non-IID and graph data, bidirectional knowledge transfer, incentive mechanism design, game mechanism design, model watermarking, and efficiency. The ten challenging problems manifest in five pivotal aspects: “Foundational Theory,” which aims to establish a coherent and unifying theoretical framework for FedFMs. “Data,” addressing the difficulties in leveraging domain-specific knowledge from private data while maintaining privacy; “Heterogeneity,” examining variations in data, model, and computational resources across clients; “Security and Privacy,” focusing on defenses against malicious attacks and model theft; and “Efficiency,” highlighting the need for improvements in training, communication, and parameter efficiency. For each problem, we offer a clear mathematical definition on the objective function, analyze existing methods, and discuss the key challenges and potential solutions. This in-depth exploration aims to advance the theoretical foundations of FedFMs, guide practical implementations, and inspire future research to overcome these obstacles, thereby enabling the robust, efficient, and privacy-preserving FedFMs in various real-world applications.

Cooperative broadcast is an efficient approach to improve Wi-Fi broadcast performance in a crowded scenario with densely deployed access points (APs). However, the current concurrent transmission MAC protocols cannot synchronize multi-APs’ signals perfectly for all users. As a result, the superimposed signal from APs is time-varying at the users due to the multiple time-domain channels and carrier frequency offsets (CFOs) from multiple APs. The traditional channel estimation approach that estimates the superimposed channel as a whole is ill-suited for the superimposed signal. In this paper, we propose a fine-grained channel estimation approach to first estimate these channel parameters for each AP, and then reconstruct the superimposed channel. Specifically, we present a two-stage channel estimation algorithm that first estimates the CFOs by discretizing the CFO range and matching the most possible CFOs, and then computes the time-domain channels. Experiment and simulation results show the new channel estimation approach achieves much lower bit error rate (BER) and packet error rate (PER) than the traditional IEEE 802.11 approach. In addition, we propose a distributed mechanism to choose the master AP that initializes multi-APs’ simultaneous transmission, which the current concurrent transmission MAC protocols lack. Network-layer simulation results show that the proposed cooperative broadcast scheme improves the throughput by 64% to 82% compared with the traditional uncooperative broadcast scheme.

Optimal execution is a sequential decision-making problem for cost-saving in algorithmic trading. Studies have found that reinforcement learning (RL) can help decide the order-splitting sizes. However, a problem remains unsolved: how to place limit orders at appropriate limit prices? The key challenge lies in the ``continuous-discrete duality'' of the action space. On the one hand, the continuous action space using percentage changes in prices is preferred for generalization. On the other hand, the trader eventually needs to choose limit prices discretely due to the existence of the tick size, which requires specialization for every single stock with different characteristics (e.g., the liquidity and the price range). So we need continuous control for generalization and discrete control for specialization. To this end, we propose a hybrid RL method to combine the advantages of both of them. We first use a continuous control agent to scope an action subset, then deploy a fine-grained agent to choose a specific limit price. Extensive experiments show that our method has higher sample efficiency and better training stability than existing RL algorithms and significantly outperforms previous learning-based methods for order execution.