Vision foundation models like the Segment Anything Model (SAM), pretrained on large-scale natural image datasets, often struggle in medical image segmentation due to a lack of domain-specific adaptation. In clinical practice, fine-tuning such models efficiently for medical downstream tasks with minimal resource demands, while maintaining strong performance, is challenging. To address these issues, we propose BALR-SAM, a boundary-aware low-rank adaptation framework that enhances SAM for medical imaging. It combines three tailored components: (1) a Complementary Detail Enhancement Network (CDEN) using depthwise separable convolutions and multi-scale fusion to capture boundary-sensitive features essential for accurate segmentation; (2) low-rank adapters integrated into SAM's Vision Transformer blocks to optimize feature representation and attention for medical contexts, while simultaneously significantly reducing the parameter space; and (3) a low-rank tensor attention mechanism in the mask decoder, cutting memory usage by 75% and boosting inference speed. Experiments on standard medical segmentation datasets show that BALR-SAM, without requiring prompts, outperforms several state-of-the-art (SOTA) methods, including fully fine-tuned MedSAM, while updating just 1.8% (11.7M) of its parameters.

High-dimensional decision-making tasks, such as business partner selection, involve evaluating large candidate pools with heterogeneous numerical, categorical, and textual features. MAS, a hierarchical multi-agent framework that decomposes evaluation into three layers: a Planner Agent that designs strategies, Specialized Agents that perform role-specific assessments, and a Supervisor Agent that integrates their outputs. To support systematic evaluation, we also introduce a curated benchmark dataset of venture capital co-investments, featuring diverse firm attributes and ground-truth syndicates. MAS consistently outperforms single-agent and debate-based multi-agent baselines, achieving up to 10-15% higher match rates. Analysis of agent reasoning shows that planners are most responsive to domain-informed prompts, specialists produce complementary feature coverage, and supervisors play an important role in aggregation. Our implementation is available at this anonymous link. In real-world decision-making, practitioners often navigate high-dimensional data including extensive option sets and numerous evaluative features (Sandanayake et al., 2018; Sigle et al., 2023). Business partner selection which includes partner shortlisting and strategic alliance formation exemplifies this challenge (Mindruta et al., 2016): firms often face a vast pool of potential candidates, each described by diverse attributes ranging from quantitative indicators (e.g., financial metrics, geographic presence) to text-rich information (e.g., strategic fit, investment preferences) (Shah & Swaminathan, 2008). The scale and complexity of such data can easily overwhelm human decision-makers, incurring significant costs (Li et al., 2008). This underscores the need for intelligent systems capable of analyzing large candidate sets and diverse features. Large language models (LLMs) have emerged as promising tools for addressing reasoning tasks in data-rich domains (Lee et al., 2025; Mischler et al., 2024). With appropriate prompting (e.g., few-shot learning) or information retrieval techniques (e.g., RAG), these models can identify salient features using only feature and task descriptions, achieving performance comparable to established methods (Li et al., 2025a; Jeong et al., 2024).

Graphical user interfaces (GUIs) are the primary medium for human-computer interaction, yet automating GUI interactions remains challenging due to the complexity of visual elements, dynamic environments, and the need for multi-step reasoning. Existing methods based on vision-language models (VLMs) often suffer from limited resolution, domain mismatch, and insufficient sequential decisionmaking capability. To address these issues, we propose Mano, a robust GUI agent built upon a multi-modal foundation model pre-trained on extensive web and computer system data. Our approach integrates a novel simulated environment for high-fidelity data generation, a three-stage training pipeline (supervised fine-tuning, offline reinforcement learning, and online reinforcement learning), and a verification module for error recovery. Mano demonstrates state-of-the-art performance on multiple GUI benchmarks, including Mind2Web and OSWorld, achieving significant improvements in success rate and operational accuracy. Our work provides new insights into the effective integration of reinforcement learning with VLMs for practical GUI agent deployment, highlighting the importance of domain-specific data, iterative training, and holistic reward design.

Discrete diffusion models have recently gained significant prominence in applications involving natural language and graph data. A key factor influencing their effectiveness is the efficiency of discretized samplers. Among these, $τ$-leaping samplers have become particularly popular due to their theoretical and empirical success. However, existing theoretical analyses of $τ$-leaping often rely on somewhat restrictive and difficult-to-verify regularity assumptions, and their convergence bounds contain quadratic dependence on the vocabulary size. In this work, we introduce a new analytical approach for discrete diffusion models that removes the need for such assumptions. For the standard $τ$-leaping method, we establish convergence guarantees in KL divergence that scale linearly with vocabulary size, improving upon prior results with quadratic dependence. Our approach is also more broadly applicable: it provides the first convergence guarantees for other widely used samplers, including the Euler method and Tweedie $τ$-leaping. Central to our approach is a novel technique based on differential inequalities, offering a more flexible alternative to the traditional Girsanov change-of-measure methods. This technique may also be of independent interest for the analysis of other stochastic processes.

This paper presents a physics-aware cyberphysical resilience framework for radial microgrids under coordinated cyberattacks. The proposed approach models the attacker through a hypergraph neural network (HGNN) enhanced with model agnostic metalearning (MAML) to rapidly adapt to evolving defense strategies and predict high-impact contingencies. The defender is modeled via a bi-level Stackelberg game, where the upper level selects optimal tie-line switching and distributed energy resource (DER) dispatch using an Alternating Direction Method of Multipliers (ADMM) coordinator embedded within the Non-dominated Sorting Genetic Algorithm II (NSGA-II). The framework simultaneously optimizes load served, operational cost, and voltage stability, ensuring all post-defense states satisfy network physics constraints. The methodology is first validated on the IEEE 69-bus distribution test system with 12 DERs, 8 critical loads, and 5 tie-lines, and then extended to higher bus systems including the IEEE 123-bus feeder and a synthetic 300-bus distribution system. Results show that the proposed defense strategy restores nearly full service for 90% of top-ranked attacks, mitigates voltage violations, and identifies Feeder 2 as the principal vulnerability corridor. Actionable operating rules are derived, recommending pre-arming of specific tie-lines to enhance resilience, while higher bus system studies confirm scalability of the framework on the IEEE 123-bus and 300-bus systems.

When online sellers use AI learning algorithms to automatically compete on e-commerce platforms, there is concern that they will learn to coordinate on higher than competitive prices. However, this concern was primarily raised in single-dimension price competition. We investigate whether this prediction holds when sellers make pricing and advertising decisions together, i.e., two-dimensional decisions. We analyze competition in multi-agent reinforcement learning, and use a large-scale dataset from Amazon.com to provide empirical evidence. We show that when consumers have high search costs, learning algorithms can coordinate on prices lower than competitive prices, facilitating a win-win-win for consumers, sellers, and platforms. This occurs because algorithms learn to coordinate on lower advertising bids, which lower advertising costs, leading to lower prices and enlarging demand on the platform. We also show that our results generalize to any learning algorithm that uses exploration of price and advertising bids. Consistent with our predictions, an empirical analysis shows that price levels exhibit a negative interaction between estimated consumer search costs and algorithm usage index. We analyze the platform's strategic response and find that reserve price adjustments will not increase platform profits, but commission adjustments will, while maintaining the beneficial outcomes for both sellers and consumers.

Multimodal large language models (MLLMs) have achieved remarkable progress across a range of vision-language tasks and demonstrate strong potential for traffic accident understanding. However, existing MLLMs in this domain primarily focus on coarse-grained image-level or video-level comprehension and often struggle to handle fine-grained visual details or localized scene components, limiting their applicability in complex accident scenarios. To address these limitations, we propose SafePLUG, a novel framework that empowers MLLMs with both Pixel-Level Understanding and temporal Grounding for comprehensive traffic accident analysis. SafePLUG supports both arbitrary-shaped visual prompts for region-aware question answering and pixel-level segmentation based on language instructions, while also enabling the recognition of temporally anchored events in traffic accident scenarios. To advance the development of MLLMs for traffic accident understanding, we curate a new dataset containing multimodal question-answer pairs centered on diverse accident scenarios, with detailed pixel-level annotations and temporal event boundaries. Experimental results show that SafePLUG achieves strong performance on multiple tasks, including region-based question answering, pixel-level segmentation, temporal event localization, and accident event understanding. These capabilities lay a foundation for fine-grained understanding of complex traffic scenes, with the potential to improve driving safety and enhance situational awareness in smart transportation systems. The code, dataset, and model checkpoints will be made publicly available at: https://zihaosheng.github.io/SafePLUG

Deep reinforcement learning has achieved impressive success in control tasks. However, its policies, represented as opaque neural networks, are often difficult for humans to understand, verify, and debug, which undermines trust and hinders real-world deployment. This work addresses this challenge by introducing a novel approach for programmatic control policy discovery, called Multimodal Large Language Model-assisted Evolutionary Search (MLES). MLES utilizes multimodal large language models as programmatic policy generators, combining them with evolutionary search to automate policy generation. It integrates visual feedback-driven behavior analysis within the policy generation process to identify failure patterns and guide targeted improvements, thereby enhancing policy discovery efficiency and producing adaptable, human-aligned policies. Experimental results demonstrate that MLES achieves performance comparable to Proximal Policy Optimization (PPO) across two standard control tasks while providing transparent control logic and traceable design processes. This approach also overcomes the limitations of predefined domain-specific languages, facilitates knowledge transfer and reuse, and is scalable across various tasks, showing promise as a new paradigm for developing transparent and verifiable control policies.

Robot teleoperation (RTo) has emerged as a viable alternative to local control, particularly when human intervention is still necessary. This research aims to study the distance effect on user perception in RTo, exploring the potential of teleoperated robots for older adult care. We propose an evaluation of non-expert users' perception of long-distance RTo, examining how their perception changes before and after interaction, as well as comparing it to that of locally operated robots. We have designed a specific protocol consisting of multiple questionnaires, along with a dedicated software architecture using the Robotics Operating System (ROS) and Unity. The results revealed no statistically significant differences between the local and remote robot conditions, suggesting that robots may be a viable alternative to traditional local control.

Trustworthy evaluations of dangerous capabilities are increasingly crucial for determining whether an AI system is safe to deploy. One empirically demonstrated threat is sandbagging - the strategic underperformance on evaluations by AI models or their developers. A promising defense is to monitor a model's chain-of-thought (CoT) reasoning, as this could reveal its intentions and plans. In this work, we measure the ability of models to sandbag on dangerous capability evaluations against a CoT monitor by prompting them to sandbag while being either monitor-oblivious or monitor-aware. We show that both frontier models and small open-sourced models can covertly sandbag against CoT monitoring 0-shot without hints. However, they cannot yet do so reliably: they bypass the monitor 16-36% of the time when monitor-aware, conditioned on sandbagging successfully. We qualitatively analyzed the uncaught CoTs to understand why the monitor failed. We reveal a rich attack surface for CoT monitoring and contribute five covert sandbagging policies generated by models. These results inform potential failure modes of CoT monitoring and may help build more diverse sandbagging model organisms.