SharpSplat: Edge-Regularized 3D Gaussian Splatting for High Fidelity Urban Building Reconstruction from UAV images

Reconstructing high-fidelity 3D building models from UAV imagery is essential for large-scale digital twin development. However, existing 3D Gaussian Splatting (3DGS) techniques often struggle with building facades, failing to capture sharp geometric transitions. To address this, we propose a semantic edge regularization framework that supervises 3DGS to produce crisp architectural boundaries. Our method leverages SAM 3 to generate precise building masks, from which we extract architecturally significant edges. During training, we align rendered image gradients with these extracted edges, forcing the Gaussians to converge into sharp structural geometries. Evaluations across campus environments, dense urban centers, and custom residential datasets demonstrate significant improvements in edge fidelity without requiring architectural modifications to the 3DGS pipeline. Our approach proves robust across diverse building types, roof geometries, and urban densities.

STAR-RIS-Assisted Integrated Sensing, Secure Communication, and Power Transfer: A Transmit Power Minimization Framework

Evolving wireless networks call for architectures that unify sensing, communication, and wireless power transfer. Although integrated sensing and communication (ISAC) and simultaneous wireless information and power transfer (SWIPT) have validated dual-function transmission, the combination of integrated sensing, secure communication, and power transfer (ISSCPT) remains largely unexplored, in part due to the tight coupling among design variables. To address this coupling and expand spatial degrees of freedom, we turn to intelligent metasurfaces: while a conventional reconfigurable intelligent surface (cRIS) reflects only to one side and thus limits coverage and flexibility, a simultaneously transmitting and reflecting RIS (STAR-RIS) enables full-space wave control, making it a natural vehicle for power-efficient ISSCPT. We study a STAR-RIS-assisted ISSCPT system and pose a central question: How much transmit power is required to operate such a system? We formulate a transmit-power minimization problem that jointly optimizes transmit and receive beamforming and the STAR-RIS configuration, and solve it via alternating optimization with successive convex approximation, second-order cone programming, and eigenvalue decomposition. Simulations show that the proposed STAR-RIS-assisted design outperforms cRIS and no-RIS baselines, and quantify the additional transmit power required by ISSCPT relative to ISAC and secure SWIPT, clarifying security-sensing-power tradeoffs in metasurface-assisted systems.

BankerToolBench: Evaluating AI Agents in End-to-End Investment Banking Workflows

Existing AI benchmarks lack the fidelity to assess economically meaningful progress on professional workflows. To evaluate frontier AI agents in a high-value, labor-intensive profession, we introduce BankerToolBench (BTB): an open-source benchmark of end-to-end analytical workflows routinely performed by junior investment bankers. To develop an ecologically valid benchmark grounded in representative work environments, we collaborated with 502 investment bankers from leading firms. BTB requires agents to execute senior banker requests by navigating data rooms, using industry tools (market data platform, SEC filings database), and generating multi-file deliverables--including Excel financial models, PowerPoint pitch decks, and PDF/Word reports. Completing a BTB task takes bankers up to 21 hours, underscoring the economic stakes of successfully delegating this work to AI. BTB enables automated evaluation of any LLM or agent, scoring deliverables against 100+ rubric criteria defined by veteran investment bankers to capture stakeholder utility. Testing 9 frontier models, we find that even the best-performing model (GPT-5.4) fails nearly half of the rubric criteria and bankers rate 0% of its outputs as client-ready. Our failure analysis reveals key obstacles (such as breakdowns in cross-artifact consistency) and improvement directions for agentic AI in high-stakes professional workflows.

Design of Uplink ISAC Systems with Cooperative Sensing: Power Control and Receive Beamforming

Integrated sensing and communication (ISAC) has emerged as a key paradigm for next-generation wireless systems, which allows wireless resources to be used for data transmission and target sensing simultaneously. In this paper, multi-user collaborative target detection in the uplink ISAC system is investigated. To incorporate the target sensing functionality, the system relies on the reuse of uplink signals from the communication users. Specifically, we analyze an uplink multi-user single-input multiple-output (MU-SIMO) communication system with bistatic sensing. Using the channel statistics, we formulate the problem of joint optimal pilot and data power allocation to maximize the uplink ergodic sum rate while meeting communication and sensing quality-of-service (QoS) requirements. To address this non-convex problem, we propose an alternating optimization (AO)-based iterative framework, where the joint power allocation problem is decomposed into two sub-problems. Specifically, the pilot power allocation is optimized using a penalty dual decomposition (PDD)-based gradient ascent algorithm, while the data power allocation is solved via successive convex approximation (SCA). Once the long-term power allocation is determined, the base station (BS) estimates the instantaneous channels using a minimum mean-squared error (MMSE) estimator. Subsequently, based on the estimated instantaneous channel state information (CSI), the receive beamforming for communication users is optimized via another SCA-based method to maximize the sum rate. Meanwhile, the optimal receive beamforming for the target is obtained in closed-form through eigenvalue decomposition (EVD).

Beyond the Limits of Rigid Arrays: Flexible Intelligent Metasurfaces for Next-Generation Wireless Networks

Following recent advances in flexible electronics and programmable metasurfaces, flexible intelligent metasurfaces (FIMs) have emerged as a promising enabling technology for next-generation wireless networks. A FIM is a morphable electromagnetic surface capable of dynamically adjusting its physical geometry to influence the radiation and propagation of electromagnetic waves. Unlike conventional rigid arrays, FIMs introduce an additional spatial degree of design freedom enabled by mechanical flexibility, which can enhance beamforming, spatial focusing, and adaptation to dynamic wireless environments. This added capability enables wireless systems to shape the propagation environment not only through electromagnetic tuning but also through controllable geometric reconfiguration. This article explores the potential of FIMs for next-generation wireless networks. We first introduce the main hardware architectures of FIMs and explain how they can be integrated into wireless communication systems. We then present representative application scenarios, highlighting the advantages of FIMs for future wireless networks and comparing them with other emerging flexible wireless technologies. To illustrate their potential impact, we present case studies comparing FIM-enabled architectures with conventional rigid-array systems, demonstrating the performance gains enabled by surface flexibility for both communication and sensing applications. Finally, we discuss key opportunities, practical challenges, and open research directions that must be addressed to fully realize the potential of FIM technology in future wireless communication systems.

Exploiting Structural Flexibility in SIM-Enabled Communications: From Adaptive Inter-Layer Spacing to Fully Morphable Layers

Stacked intelligent metasurfaces (SIMs) have recently emerged as a promising metasurface-based physical-layer paradigm for wireless communications, enabling wave-domain signal processing through multiple cascaded metasurface layers. However, conventional SIM designs rely on rigid planar layers with fixed interlayer spacing, which constrain the propagation geometry and can lead to performance saturation as the number of layers increases. This paper investigates the potential of introducing structural flexibility into SIM-enabled communication systems. Specifically, we consider two flexible SIM architectures: distance-adaptive SIM (DSIM), where interlayer distances are optimized, and stacked flexible intelligent metasurface (SFIM), where each metasurface layer is fully morphable. We jointly design the meta-atom positions and responses together with the transmit beamformer to maximize the system sum rate under per-user rate, quantization, morphing, and interlayer distance constraints. An alternating optimization framework combining gradient projection, penalty-based method, and successive convex approximation is developed to address the resulting non-convex problems. Perturbation analysis reveals that the flexibility gains of both DSIM and SFIM scale approximately linearly with the morphing range, with SFIM exhibiting a faster growth rate. Simulation results demonstrate that flexible SIM designs mitigate performance saturation with increasing layers and achieve significant transmit power savings compared to rigid SIMs.

Flexible Intelligent Metasurface for Downlink Communications under Statistical CSI

Flexible intelligent metasurface (FIM) is a recently developed, groundbreaking hardware technology with promising potential for 6G wireless systems. Unlike conventional rigid antenna array (RAA)-based transmitters, FIM-assisted transmitters can dynamically alter their physical surface through morphing, offering new degrees of freedom to enhance system performance. In this letter, we depart from prior works that rely on instantaneous channel state information (CSI) and instead address the problem of average sum spectral efficiency maximization under statistical CSI in a FIM-assisted downlink multiuser multiple-input single-output setting. To this end, we first derive the spatial correlation matrix for the FIM-aided transmitter and then propose an iterative FIM optimization algorithm based on the gradient projection method. Simulation results show that with statistical CSI, the FIM-aided system provides a significant performance gain over its RAA-based counterpart in scenarios with strong spatial channel correlation, whereas the gain diminishes when the channels are weakly correlated.

Hiding in Plain Sight: RIS-Aided Target Obfuscation in ISAC

Integrated sensing and communication (ISAC) has been identified as a promising technology for the sixth generation (6G) of communication networks. Target privacy in ISAC is essential to ensure that only legitimate sensors can detect the target while keeping it hidden from malicious ones. In this paper, we consider a downlink reconfigurable intelligent surface (RIS)-assisted ISAC system capable of protecting a sensing region against an adversarial detector. The RIS consists of both reflecting and sensing elements, adaptively changing the element assignment based on system needs. To achieve this, we minimize the maximum sensing signal-to-interference-plus-noise-ratio (SINR) at the adversarial detector within sample points in the sensing region, by optimizing the transmit beamformer at the base station, the RIS phase shift matrix, the received beamformer at the RIS, and the division between reflecting and absorptive elements at the RIS, where the latter function as sensing elements. At the same time, the system is designed to maintain a minimum sensing SINR at each monitored location, as well as minimum communication SINR for each user. To solve this challenging optimization problem, we develop an alternating optimization approach combined with a successive convex approximation based method tailored for each subproblem. Our results show that the proposed approach achieves a 25 dB reduction in the maximum sensing SINR at the adversarial detector compared to scenarios without sensing area protection. Also, the optimal RIS element assignment can further improve sensing protection by 3 dB over RISs with fixed element configuration.

Beamforming Design for Secure RIS-Enabled ISAC: Passive RIS vs. Active RIS

The forthcoming sixth-generation (6G) communications standard is anticipated to provide integrated sensing and communication (ISAC) as a fundamental service. These ISAC systems present unique security challenges because of the exposure of information-bearing signals to sensing targets, enabling them to potentially eavesdrop on sensitive communication information with the assistance of sophisticated receivers. Recently, reconfigurable intelligent surfaces (RISs) have shown promising results in enhancing the physical layer security of various wireless communication systems, including ISAC. However, the performance of conventional passive RIS (pRIS)-enabled systems are often limited due to multiplicative fading, which can be alleviated using active RIS (aIRS). In this paper, we consider the problem of beampattern gain maximization in a secure pRIS/aRIS-enabled ISAC system, subject to signal-to-interference-plus-noise ratio constraints at communication receivers, and information leakage constraints at an eavesdropping target. For the challenging non-convex problem of joint beamforming design at the base station and the pRIS/aRIS, we propose a novel successive convex approximation (SCA)-based method. Unlike the conventional alternating optimization (AO)-based methods, in the proposed SCA-based approach, all of the optimization variables are updated simultaneously in each iteration. The proposed method shows significant performance superiority for pRIS-aided ISAC system compared to a benchmark scheme using penalty-based AO method. Moreover, our simulation results also confirm that aRIS-aided system has a notably higher beampattern gain at the target compared to that offered by the pRIS-aided system for the same power budget. We also present a detailed complexity analysis and proof of convergence for the proposed SCA-based method.

PAFFA: Premeditated Actions For Fast Agents

Modern AI assistants have made significant progress in natural language understanding and API/tool integration, with emerging efforts to incorporate diverse interfaces (such as Web interfaces) for enhanced scalability and functionality. However, current approaches that heavily rely on repeated LLM-driven HTML parsing are computationally expensive and error-prone, particularly when handling dynamic web interfaces and multi-step tasks. To overcome these challenges, we introduce PAFFA (Premeditated Actions For Fast Agents), a framework designed to enhance web interaction capabilities through an Action API Library of reusable, verified browser interaction functions. By pre-computing interaction patterns and employing two core methodologies - "Dist-Map" for task-agnostic element distillation and "Unravel" for incremental page-wise exploration - PAFFA reduces inference calls by 87% while maintaining robust performance even as website structures evolve. This framework accelerates multi-page task execution and offers a scalable solution to advance autonomous web agent research.