Multi-view understanding, the ability to reconcile visual information across diverse viewpoints for effective navigation, manipulation, and 3D scene comprehension, is a fundamental challenge in Multi-Modal Large Language Models (MLLMs) to be used as embodied agents. While recent MLLMs have shown impressive advances in high-level reasoning and planning, they frequently fall short when confronted with multi-view geometric consistency and cross-view correspondence. To comprehensively evaluate the challenges of MLLMs in multi-view scene reasoning, we propose All-Angles Bench, a benchmark of over 2,100 human carefully annotated multi-view question-answer pairs across 90 diverse real-world scenes. Our six tasks (counting, attribute identification, relative distance, relative direction, object manipulation, and camera pose estimation) specifically test model's geometric correspondence and the capacity to align information consistently across views. Our extensive experiments, benchmark on 27 representative MLLMs including Gemini-2.0-Flash, Claude-3.7-Sonnet, and GPT-4o against human evaluators reveals a substantial performance gap, indicating that current MLLMs remain far from human-level proficiency. Through in-depth analysis, we show that MLLMs are particularly underperforming under two aspects: (1) cross-view correspondence for partially occluded views and (2) establishing the coarse camera poses. These findings highlight the necessity of domain-specific refinements or modules that embed stronger multi-view awareness. We believe that our All-Angles Bench offers valuable insights and contribute to bridging the gap between MLLMs and human-level multi-view understanding. The project and benchmark are publicly available at https://danielchyeh.github.io/All-Angles-Bench/.
We design a suite of minimal algorithmic tasks that are a loose abstraction of open-ended real-world tasks. This allows us to cleanly and controllably quantify the creative limits of the present-day language model. Much like real-world tasks that require a creative, far-sighted leap of thought, our tasks require an implicit, open-ended stochastic planning step that either (a) discovers new connections in an abstract knowledge graph (like in wordplay, drawing analogies, or research) or (b) constructs new patterns (like in designing math problems or new proteins). In these tasks, we empirically and conceptually argue how next-token learning is myopic and memorizes excessively; comparatively, multi-token approaches, namely teacherless training and diffusion models, excel in producing diverse and original output. Secondly, in our tasks, we find that to elicit randomness from the Transformer without hurting coherence, it is better to inject noise right at the input layer (via a method we dub hash-conditioning) rather than defer to temperature sampling from the output layer. Thus, our work offers a principled, minimal test-bed for analyzing open-ended creative skills, and offers new arguments for going beyond next-token learning and softmax-based sampling. We make part of the code available under https://github.com/chenwu98/algorithmic-creativity
Modern power systems face growing risks from cyber-physical attacks, necessitating enhanced resilience due to their societal function as critical infrastructures. The challenge is that defense of large-scale systems-of-systems requires scalability in their threat and risk assessment environment for cyber physical analysis including cyber-informed transmission planning, decision-making, and intrusion response. Hence, we present a scalable discrete event simulation tool for analysis of energy systems, called DESTinE. The tool is tailored for largescale cyber-physical systems, with a focus on power systems. It supports faster-than-real-time traffic generation and models packet flow and congestion under both normal and adversarial conditions. Using three well-established power system synthetic cases with 500, 2000, and 10,000 buses, we overlay a constructed cyber network employing star and radial topologies. Experiments are conducted to identify critical nodes within a communication network in response to a disturbance. The findings are incorporated into a constrained optimization problem to assess the impact of the disturbance on a specific node and its cascading effects on the overall network. Based on the solution of the optimization problem, a new hybrid network topology is also derived, combining the strengths of star and radial structures to improve network resilience. Furthermore, DESTinE is integrated with a virtual server and a hardware-in-the-loop (HIL) system using Raspberry Pi 5.
Multi-Agent Pickup and Delivery (MAPD) is a fundamental problem in robotics, particularly in applications such as warehouse automation and logistics. Existing solutions often face challenges in scalability, adaptability, and efficiency, limiting their applicability in dynamic environments with real-time planning requirements. This paper presents Neural ATTF (Adaptive Task Token Framework), a new algorithm that combines a Priority Guided Task Matching (PGTM) Module with Neural STA* (Space-Time A*), a data-driven path planning method. Neural STA* enhances path planning by enabling rapid exploration of the search space through guided learned heuristics and ensures collision avoidance under dynamic constraints. PGTM prioritizes delayed agents and dynamically assigns tasks by prioritizing agents nearest to these tasks, optimizing both continuity and system throughput. Experimental evaluations against state-of-the-art MAPD algorithms, including TPTS, CENTRAL, RMCA, LNS-PBS, and LNS-wPBS, demonstrate the superior scalability, solution quality, and computational efficiency of Neural ATTF. These results highlight the framework's potential for addressing the critical demands of complex, real-world multi-agent systems operating in high-demand, unpredictable settings.
Deploying robot learning methods to a quadrotor in unstructured outdoor environments is an exciting task. Quadrotors operating in real-world environments by learning-based methods encounter several challenges: a large amount of simulator generated data required for training, strict demands for real-time processing onboard, and the sim-to-real gap caused by dynamic and noisy conditions. Current works have made a great breakthrough in applying learning-based methods to end-to-end control of quadrotors, but rarely mention the infrastructure system training from scratch and deploying to reality, which makes it difficult to reproduce methods and applications. To bridge this gap, we propose a platform that enables the seamless transfer of end-to-end deep reinforcement learning (DRL) policies. We integrate the training environment, flight dynamics control, DRL algorithms, the MAVROS middleware stack, and hardware into a comprehensive workflow and architecture that enables quadrotors' policies to be trained from scratch to real-world deployment in several minutes. Our platform provides rich types of environments including hovering, dynamic obstacle avoidance, trajectory tracking, balloon hitting, and planning in unknown environments, as a physical experiment benchmark. Through extensive empirical validation, we demonstrate the efficiency of proposed sim-to-real platform, and robust outdoor flight performance under real-world perturbations. Details can be found from our website https://emnavi.tech/AirGym/.
In interactions between automated vehicles (AVs) and crossing pedestrians, modeling implicit vehicle communication is crucial. In this work, we present a combined prediction and planning approach that allows to consider the influence of the planned vehicle behavior on a pedestrian and predict a pedestrian's reaction. We plan the behavior by solving two consecutive optimal control problems (OCPs) analytically, using variational calculus. We perform a validation step that assesses whether the planned vehicle behavior is adequate to trigger a certain pedestrian reaction, which accounts for the closed-loop characteristics of prediction and planning influencing each other. In this step, we model the influence of the planned vehicle behavior on the pedestrian using a probabilistic behavior acceptance model that returns an estimate for the crossing probability. The probabilistic modeling of the pedestrian reaction facilitates considering the pedestrian's costs, thereby improving cooperative behavior planning. We demonstrate the performance of the proposed approach in simulated vehicle-pedestrian interactions with varying initial settings and highlight the decision making capabilities of the planning approach.
A new class of Multi-Rotor Aerial Vehicles (MRAVs), known as omnidirectional MRAVs (o-MRAVs), has gained attention for their ability to independently control 3D position and orientation. This capability enhances robust planning and control in aerial communication networks, enabling more adaptive trajectory planning and precise antenna alignment without additional mechanical components. These features are particularly valuable in uncertain environments, where disturbances such as wind and interference affect communication stability. This paper examines o-MRAVs in the context of robust aerial network planning, comparing them with the more common under-actuated MRAVs (u-MRAVs). Key applications, including physical layer security, optical communications, and network densification, are highlighted, demonstrating the potential of o-MRAVs to improve reliability and efficiency in dynamic communication scenarios.
We address the problem of planning under uncertainty, where an agent must choose actions that not only achieve desired outcomes but also reduce uncertainty. Traditional methods often treat exploration and exploitation as separate objectives, lacking a unified inferential foundation. Active inference, grounded in the Free Energy Principle, offers such a foundation by minimizing Expected Free Energy (EFE), a cost function that combines utility with epistemic drives like ambiguity resolution and novelty seeking. However, the computational burden of EFE minimization has remained a major obstacle to its scalability. In this paper, we show that EFE-based planning arises naturally from minimizing a variational free energy functional on a generative model augmented with preference and epistemic priors. This result reinforces theoretical consistency with the Free Energy Principle, by casting planning itself as variational inference. Our formulation yields optimal policies that jointly support goal achievement and information gain, while incorporating a complexity term that accounts for bounded computational resources. This unifying framework connects and extends existing methods, enabling scalable, resource-aware implementations of active inference agents.
This paper develops a novel unmanned surface vehicle (USV)-autonomous underwater vehicle (AUV) collaborative system designed to enhance underwater task performance in extreme sea conditions. The system integrates a dual strategy: (1) high-precision multi-AUV localization enabled by Fisher information matrix-optimized USV path planning, and (2) reinforcement learning-based cooperative planning and control method for multi-AUV task execution. Extensive experimental evaluations in the underwater data collection task demonstrate the system's operational feasibility, with quantitative results showing significant performance improvements over baseline methods. The proposed system exhibits robust coordination capabilities between USV and AUVs while maintaining stability in extreme sea conditions. To facilitate reproducibility and community advancement, we provide an open-source simulation toolkit available at: https://github.com/360ZMEM/USV-AUV-colab .
Communication is fundamental for multi-robot collaboration, with accurate radio mapping playing a crucial role in predicting signal strength between robots. However, modeling radio signal propagation in large and occluded environments is challenging due to complex interactions between signals and obstacles. Existing methods face two key limitations: they struggle to predict signal strength for transmitter-receiver pairs not present in the training set, while also requiring extensive manual data collection for modeling, making them impractical for large, obstacle-rich scenarios. To overcome these limitations, we propose FERMI, a flexible radio mapping framework. FERMI combines physics-based modeling of direct signal paths with a neural network to capture environmental interactions with radio signals. This hybrid model learns radio signal propagation more efficiently, requiring only sparse training data. Additionally, FERMI introduces a scalable planning method for autonomous data collection using a multi-robot team. By increasing parallelism in data collection and minimizing robot travel costs between regions, overall data collection efficiency is significantly improved. Experiments in both simulation and real-world scenarios demonstrate that FERMI enables accurate signal prediction and generalizes well to unseen positions in complex environments. It also supports fully autonomous data collection and scales to different team sizes, offering a flexible solution for creating radio maps. Our code is open-sourced at https://github.com/ymLuo1214/Flexible-Radio-Mapping.
Hierarchical composite endpoints, such as those analyzed using the Finkelstein-Schoenfeld (FS) statistic, are increasingly used in clinical trials for their ability to incorporate clinically prioritized outcomes. However, adaptive design methods for these endpoints remain underdeveloped. This paper presents a practical framework for implementing sample size re-estimation (SSR) in trials using hierarchical composites, motivated by a cardiovascular trial with mortality, hospitalization, and a functional response as prioritized endpoints. We use a two-stage adaptive design with a single interim analysis for illustration. The interim analysis incorporates predictive probabilities to determine whether the trial should stop for futility, continue as planned, or increase the sample size to maintain power. The decision framework is based on predefined zones for predictive probability, with corresponding adjustments to the stage 2 sample size. Simulation studies across various treatment scenarios demonstrate strong type I error control and increased power compared to a fixed design, particularly for treatment effects that are clinically relevant but lower than the alternative hypothesis. We also explore an alternative conditional power approach for SSR, offering further sample size optimization. Our results support the use of SSR with hierarchical composite outcomes using an FS statistic, enhancing trial efficiency.
Autonomous driving has garnered significant attention for its potential to improve safety, traffic efficiency, and user convenience. However, the dynamic and complex nature of interactive driving poses significant challenges, including the need to navigate non-linear road geometries, handle dynamic obstacles, and meet stringent safety and comfort requirements. Traditional approaches, such as artificial potential fields (APF), often fall short in addressing these complexities independently, necessitating the development of integrated and adaptive frameworks. This paper presents a novel approach to autonomous vehicle navigation that integrates artificial potential fields, Frenet coordinates, and improved particle swarm optimization (IPSO). A dynamic risk field, adapted from traditional APF, is proposed to ensure interactive safety by quantifying risks and dynamically adjusting lane-changing intentions based on surrounding vehicle behavior. Frenet coordinates are utilized to simplify trajectory planning on non-straight roads, while an enhanced quintic polynomial trajectory generator ensures smooth and comfortable path transitions. Additionally, an IPSO algorithm optimizes trajectory selection in real time, balancing safety and user comfort within a feasible input range. The proposed framework is validated through extensive simulations and real-world scenarios, demonstrating its ability to navigate complex traffic environments, maintain safety margins, and generate smooth, dynamically feasible trajectories.
Accurate and efficient medical image segmentation is crucial for advancing clinical diagnostics and surgical planning, yet remains a complex challenge due to the variability in anatomical structures and the demand for low-complexity models. In this paper, we introduced Med-2D SegNet, a novel and highly efficient segmentation architecture that delivers outstanding accuracy while maintaining a minimal computational footprint. Med-2D SegNet achieves state-of-the-art performance across multiple benchmark datasets, including KVASIR-SEG, PH2, EndoVis, and GLAS, with an average Dice similarity coefficient (DSC) of 89.77% across 20 diverse datasets. Central to its success is the compact Med Block, a specialized encoder design that incorporates dimension expansion and parameter reduction, enabling precise feature extraction while keeping model parameters to a low count of just 2.07 million. Med-2D SegNet excels in cross-dataset generalization, particularly in polyp segmentation, where it was trained on KVASIR-SEG and showed strong performance on unseen datasets, demonstrating its robustness in zero-shot learning scenarios, even though we acknowledge that further improvements are possible. With top-tier performance in both binary and multi-class segmentation, Med-2D SegNet redefines the balance between accuracy and efficiency, setting a new benchmark for medical image analysis. This work paves the way for developing accessible, high-performance diagnostic tools suitable for clinical environments and resource-constrained settings, making it a step forward in the democratization of advanced medical technology.
Machine learning (ML)-based planners have recently gained significant attention. They offer advantages over traditional optimization-based planning algorithms. These advantages include fewer manually selected parameters and faster development. Within ML-based planning, imitation learning (IL) is a common algorithm. It primarily learns driving policies directly from supervised trajectory data. While IL has demonstrated strong performance on many open-loop benchmarks, it remains challenging to determine if the learned policy truly understands fundamental driving principles, rather than simply extrapolating from the ego-vehicle's initial state. Several studies have identified this limitation and proposed algorithms to address it. However, these methods often use original datasets for evaluation. In these datasets, future trajectories are heavily dependent on initial conditions. Furthermore, IL often overfits to the most common scenarios. It struggles to generalize to rare or unseen situations. To address these challenges, this work proposes: 1) a novel closed-loop simulator supporting both imitation and reinforcement learning, 2) a causal benchmark derived from the Waymo Open Dataset to rigorously assess the impact of the copycat problem, and 3) a novel framework integrating imitation learning and reinforcement learning to overcome the limitations of purely imitative approaches. The code for this work will be released soon.
The moving target traveling salesman problem with obstacles (MT-TSP-O) seeks an obstacle-free trajectory for an agent that intercepts a given set of moving targets, each within specified time windows, and returns to the agent's starting position. Each target moves with a constant velocity within its time windows, and the agent has a speed limit no smaller than any target's speed. We present FMC*-TSP, the first complete and bounded-suboptimal algorithm for the MT-TSP-O, and results for an agent whose configuration space is $\mathbb{R}^3$. Our algorithm interleaves a high-level search and a low-level search, where the high-level search solves a generalized traveling salesman problem with time windows (GTSP-TW) to find a sequence of targets and corresponding time windows for the agent to visit. Given such a sequence, the low-level search then finds an associated agent trajectory. To solve the low-level planning problem, we develop a new algorithm called FMC*, which finds a shortest path on a graph of convex sets (GCS) via implicit graph search and pruning techniques specialized for problems with moving targets. We test FMC*-TSP on 280 problem instances with up to 40 targets and demonstrate its smaller median runtime than a baseline based on prior work.
Traditional urban planning methodologies often fail to capture the complexity of contemporary urbanization and environmental sustainability challenges. This study investigates the integration of Generative Design, Virtual Reality (VR), and Digital Twins (DT) to enhance walkability in urban planning. VR provides distinct benefits over conventional approaches, including 2D maps, static renderings, and physical models, by allowing stakeholders to engage with urban designs more intuitively, identify walkability challenges, and suggest iterative improvements. Preliminary findings from structured interviews with Eindhoven residents provide critical insights into pedestrian preferences and walkability considerations. The next phase of the study involves the development of VR-DT integrated prototypes to simulate urban environments, assess walkability, and explore the role of Generative Design in generating adaptive urban planning solutions. The objective is to develop a decision-support tool that enables urban planners to incorporate diverse stakeholder perspectives, optimize pedestrian-oriented urban design, and advance regenerative development principles. By leveraging these emerging technologies, this research contributes to the evolution of data-driven, participatory urban planning frameworks aimed at fostering sustainable and walkable cities.
Satellite-based communication systems are integral to delivering high-speed data services in aviation, particularly for business aviation operations requiring global connectivity. These systems, however, are challenged by a multitude of interdependent factors such as satellite handovers, congestion, flight maneuvers and seasonal trends, making network performance prediction a complex task. No established methodologies currently exist for network performance prediction in avionic communication systems. This paper addresses the gap by proposing machine learning (ML)-based approaches for pre-flight network performance predictions. The proposed models predict performance along a given flight path, taking as input positional and network-related information and outputting the predicted performance for each position. In business aviation, flight crews typically have multiple flight plans to choose from for each city pair, allowing them to select the most optimal option. This approach enables proactive decision-making, such as selecting optimal flight paths prior to departure.
In automated web testing, generating test scripts from natural language task descriptions is crucial for enhancing the test generation process. This activity involves creating the correct sequences of actions to form test scripts for future testing activities. Current state-of-the-art approaches are limited in generating these action sequences, as they either demand substantial manual effort for human demonstrations or fail to consider the history of previous web content and actions to decide the next action. In this paper, we introduce HxAgent, an iterative large language model agent planning approach that determines the next action based on: 1) observations of the current contents and feasible actions, 2) short-term memory of previous web states and actions, and 3) long-term experience with (in)correct action sequences. The agent generates a sequence of actions to perform a given task, which is effectively an automated test case to verify the task. We conducted an extensive empirical evaluation of HxAgent using two datasets. On the MiniWoB++ dataset, our approach achieves 97% exact-match accuracy that is comparable to the best baselines while eliminating the need for human demonstrations required by those methods. For complex tasks requiring navigation through multiple actions and screens, HxAgent achieves an average 82% exact-match. On the second dataset, comprising 350 task instances across seven popular websites, including YouTube, LinkedIn, Facebook, and Google, HxAgent achieves high performance, with 87% of the action sequences exactly matching the ground truth and a prefix-match of 93%, outperforming the baseline by 59%.
The scarcity of high-quality residential load data can pose obstacles for decarbonizing the residential sector as well as effective grid planning and operation. The above challenges have motivated research into generating synthetic load data, but existing methods faced limitations in terms of scalability, diversity, and similarity. This paper proposes a Generative Adversarial Network-based Synthetic Residential Load Pattern (RLP-GAN) generation model, a novel weakly-supervised GAN framework, leveraging an over-complete autoencoder to capture dependencies within complex and diverse load patterns and learn household-level data distribution at scale. We incorporate a model weight selection method to address the mode collapse problem and generate load patterns with high diversity. We develop a holistic evaluation method to validate the effectiveness of RLP-GAN using real-world data of 417 households. The results demonstrate that RLP-GAN outperforms state-of-the-art models in capturing temporal dependencies and generating load patterns with higher similarity to real data. Furthermore, we have publicly released the RLP-GAN generated synthetic dataset, which comprises one million synthetic residential load pattern profiles.
Data-driven emulation of nonlinear dynamics is challenging due to long-range skill decay that often produces physically unrealistic outputs. Recent advances in generative modeling aim to address these issues by providing uncertainty quantification and correction. However, the quality of generated simulation remains heavily dependent on the choice of conditioning priors. In this work, we present an efficient generative framework for dynamics emulation, unifying principles of turbulence with diffusion-based modeling: Cohesion. Specifically, our method estimates large-scale coherent structure of the underlying dynamics as guidance during the denoising process, where small-scale fluctuation in the flow is then resolved. These coherent priors are efficiently approximated using reduced-order models, such as deep Koopman operators, that allow for rapid generation of long prior sequences while maintaining stability over extended forecasting horizon. With this gain, we can reframe forecasting as trajectory planning, a common task in reinforcement learning, where conditional denoising is performed once over entire sequences, minimizing the computational cost of autoregressive-based generative methods. Empirical evaluations on chaotic systems of increasing complexity, including Kolmogorov flow, shallow water equations, and subseasonal-to-seasonal climate dynamics, demonstrate Cohesion superior long-range forecasting skill that can efficiently generate physically-consistent simulations, even in the presence of partially-observed guidance.
The requirement for autonomous robots to exhibit higher-level cognitive skills by planning and adapting in an ever-changing environment is indeed a great challenge for the AI community. Progress has been made in the automated planning community on refinement and repair of an agent's symbolic knowledge to do task planning in an incomplete or changing environmental model, but these advances up to now have not been transferred to real physical robots. This paper demonstrates how a physical robot can be capable of adapting its symbolic knowledge of the environment, by using experiences in robot action execution to drive knowledge refinement and hence to improve the success rate of the task plans the robot creates. To implement more robust planning systems, we propose a method for refining domain knowledge to improve the knowledge on which intelligent robot behavior is based. This architecture has been implemented and evaluated using a NAO robot. The refined knowledge leads to the future synthesis of task plans which demonstrate decreasing rates of failure over time as faulty knowledge is removed or adjusted.
Traffic flow prediction plays an important role in Intelligent Transportation Systems in traffic management and urban planning. There have been extensive successful works in this area. However, these approaches focus only on modelling the flow transition and ignore the flow generation process, which manifests itself in two ways: (i) The models are based on Markovian assumptions, ignoring the multi-periodicity of the flow generation in nodes. (ii) The same structure is designed to encode both the transition and generation processes, ignoring the differences between them. To address these problems, we propose an Effective Multi-Branch Similarity Transformer for Traffic Flow Prediction, namely EMBSFormer. Through data analysis, we find that the factors affecting traffic flow include node-level traffic generation and graph-level traffic transition, which describe the multi-periodicity and interaction pattern of nodes, respectively. Specifically, to capture traffic generation patterns, we propose a similarity analysis module that supports multi-branch encoding to dynamically expand significant cycles. For traffic transition, we employ a temporal and spatial self-attention mechanism to maintain global node interactions, and use GNN and time conv to model local node interactions, respectively. Model performance is evaluated on three real-world datasets on both long-term and short-term prediction tasks. Experimental results show that EMBSFormer outperforms baselines on both tasks. Moreover, compared to models based on flow transition modelling (e.g. GMAN, 513k), the variant of EMBSFormer(93K) only uses 18\% of the parameters, achieving the same performance.
Multimodal Large Language Models (MLLMs) have powered Graphical User Interface (GUI) Agents, showing promise in automating tasks on computing devices. Recent works have begun exploring reasoning in GUI tasks with encouraging results. However, many current approaches rely on manually designed reasoning templates, which may result in reasoning that is not sufficiently robust and adaptive for complex GUI environments. Meanwhile, some existing agents continue to operate as Reactive Actors, relying primarily on implicit reasoning that may lack sufficient depth for GUI tasks demanding planning and error recovery. We argue that advancing these agents requires a shift from reactive acting towards acting based on deliberate reasoning. To facilitate this transformation, we introduce InfiGUI-R1, an MLLM-based GUI agent developed through our Actor2Reasoner framework, a reasoning-centric, two-stage training approach designed to progressively evolve agents from Reactive Actors to Deliberative Reasoners. The first stage, Reasoning Injection, focuses on establishing a basic reasoner. We employ Spatial Reasoning Distillation to transfer cross-modal spatial reasoning capabilities from teacher models to MLLMs through trajectories with explicit reasoning steps, enabling models to integrate GUI visual-spatial information with logical reasoning before action generation. The second stage, Deliberation Enhancement, refines the basic reasoner into a deliberative one using Reinforcement Learning. This stage introduces two approaches: Sub-goal Guidance, which rewards models for generating accurate intermediate sub-goals, and Error Recovery Scenario Construction, which creates failure-and-recovery training scenarios from identified prone-to-error steps. Experimental results show InfiGUI-R1 achieves strong performance in GUI grounding and trajectory tasks. Resources at https://github.com/Reallm-Labs/InfiGUI-R1.
Robust optimization is a commonly employed method to mitigate uncertainties in the planning of intensity-modulated proton therapy (IMPT). In certain contexts, the large number of uncertainty scenarios makes the robust problem impractically expensive to solve. Recent developments in research on IMPT treatment planning indicate that the number of ideally considered error scenarios may continue to increase. In this paper, we therefore investigate methods that reduce the size of the scenario set considered during the robust optimization. Six cases of patients with non-small cell lung cancer are considered. First, we investigate the existence of an optimal subset of scenarios that needs to be considered during robust optimization, and perform experiments to see if the set can be found in a reasonable time and substitute for the full set of scenarios during robust IMPT optimization. We then consider heuristic methods to estimate this subset or find subsets with similar properties. Specifically, we select a subset of maximal diversity in terms of scenario-specific features such as the dose distributions and function gradients at the initial point. Finally, we consider adversarial methods as an alternative to solving the full robust problem and investigate the impact on computation times. The results indicate that the optimal subset can be well approximated by solving the robust IMPT problem with conventional methods. Of the methods designed to approximate it within a practically useful time frame, the results of the diversity-maximization methods indicate that they may perform better than a manual selection of scenarios based on the patient geometry. In addition, the adversarial approaches decreased the computation time by at least half compared to the conventional approach.
Image-based crack detection algorithms are increasingly in demand in infrastructure monitoring, as early detection of cracks is of paramount importance for timely maintenance planning. While deep learning has significantly advanced crack detection algorithms, existing models often require extensive labeled datasets and high computational costs for fine-tuning, limiting their adaptability across diverse conditions. This study introduces an efficient selective fine-tuning strategy, focusing on tuning normalization components, to enhance the adaptability of segmentation models for crack detection. The proposed method is applied to the Segment Anything Model (SAM) and five well-established segmentation models. Experimental results demonstrate that selective fine-tuning of only normalization parameters outperforms full fine-tuning and other common fine-tuning techniques in both performance and computational efficiency, while improving generalization. The proposed approach yields a SAM-based model, Segment Any Crack (SAC), achieving a 61.22\% F1-score and 44.13\% IoU on the OmniCrack30k benchmark dataset, along with the highest performance across three zero-shot datasets and the lowest standard deviation. The results highlight the effectiveness of the adaptation approach in improving segmentation accuracy while significantly reducing computational overhead.
The undergoing energy transition is causing behavioral changes in electricity use, e.g. with self-consumption of local generation, or flexibility services for demand control. To better understand these changes and the challenges they induce, accessing individual smart meter data is crucial. Yet this is personal data under the European GDPR. A widespread use of such data requires thus to create synthetic realistic and privacy-preserving samples. This paper introduces a new synthetic load curve dataset generated by conditional latent diffusion. We also provide the contracted power, time-of-use plan and local temperature used for generation. Fidelity, utility and privacy of the dataset are thoroughly evaluated, demonstrating its good quality and thereby supporting its interest for energy modeling applications.
Knitting, a cornerstone of textile manufacturing, is uniquely challenging to automate, particularly in terms of converting fabric designs into precise, machine-readable instructions. This research bridges the gap between textile production and robotic automation by proposing a novel deep learning-based pipeline for reverse knitting to integrate vision-based robotic systems into textile manufacturing. The pipeline employs a two-stage architecture, enabling robots to first identify front labels before inferring complete labels, ensuring accurate, scalable pattern generation. By incorporating diverse yarn structures, including single-yarn (sj) and multi-yarn (mj) patterns, this study demonstrates how our system can adapt to varying material complexities. Critical challenges in robotic textile manipulation, such as label imbalance, underrepresented stitch types, and the need for fine-grained control, are addressed by leveraging specialized deep-learning architectures. This work establishes a foundation for fully automated robotic knitting systems, enabling customizable, flexible production processes that integrate perception, planning, and actuation, thereby advancing textile manufacturing through intelligent robotic automation.
In automated driving, predicting trajectories of surrounding vehicles supports reasoning about scene dynamics and enables safe planning for the ego vehicle. However, existing models handle predictions as an instantaneous task of forecasting future trajectories based on observed information. As time proceeds, the next prediction is made independently of the previous one, which means that the model cannot correct its errors during inference and will repeat them. To alleviate this problem and better leverage temporal data, we propose a novel retrospection technique. Through training on closed-loop rollouts the model learns to use aggregated feedback. Given new observations it reflects on previous predictions and analyzes its errors to improve the quality of subsequent predictions. Thus, the model can learn to correct systematic errors during inference. Comprehensive experiments on nuScenes and Argoverse demonstrate a considerable decrease in minimum Average Displacement Error of up to 31.9% compared to the state-of-the-art baseline without retrospection. We further showcase the robustness of our technique by demonstrating a better handling of out-of-distribution scenarios with undetected road-users.
Change detection (CD) in remote sensing imagery plays a crucial role in various applications such as urban planning, damage assessment, and resource management. While deep learning approaches have significantly advanced CD performance, current methods suffer from poor domain adaptability, requiring extensive labeled data for retraining when applied to new scenarios. This limitation severely restricts their practical applications across different datasets. In this work, we propose DAM-Net: a Domain Adaptation Network with Micro-Labeled Fine-Tuning for CD. Our network introduces adversarial domain adaptation to CD for, utilizing a specially designed segmentation-discriminator and alternating training strategy to enable effective transfer between domains. Additionally, we propose a novel Micro-Labeled Fine-Tuning approach that strategically selects and labels a minimal amount of samples (less than 1%) to enhance domain adaptation. The network incorporates a Multi-Temporal Transformer for feature fusion and optimized backbone structure based on previous research. Experiments conducted on the LEVIR-CD and WHU-CD datasets demonstrate that DAM-Net significantly outperforms existing domain adaptation methods, achieving comparable performance to semi-supervised approaches that require 10% labeled data while using only 0.3% labeled samples. Our approach significantly advances cross-dataset CD applications and provides a new paradigm for efficient domain adaptation in remote sensing. The source code of DAM-Net will be made publicly available upon publication.
Recent advancements in dialogue policy planning have emphasized optimizing system agent policies to achieve predefined goals, focusing on strategy design, trajectory acquisition, and efficient training paradigms. However, these approaches often overlook the critical role of user characteristics, which are essential in real-world scenarios like conversational search and recommendation, where interactions must adapt to individual user traits such as personality, preferences, and goals. To address this gap, we first conduct a comprehensive study utilizing task-specific user personas to systematically assess dialogue policy planning under diverse user behaviors. By leveraging realistic user profiles for different tasks, our study reveals significant limitations in existing approaches, highlighting the need for user-tailored dialogue policy planning. Building on this foundation, we present the User-Tailored Dialogue Policy Planning (UDP) framework, which incorporates an Intrinsic User World Model to model user traits and feedback. UDP operates in three stages: (1) User Persona Portraying, using a diffusion model to dynamically infer user profiles; (2) User Feedback Anticipating, leveraging a Brownian Bridge-inspired anticipator to predict user reactions; and (3) User-Tailored Policy Planning, integrating these insights to optimize response strategies. To ensure robust performance, we further propose an active learning approach that prioritizes challenging user personas during training. Comprehensive experiments on benchmarks, including collaborative and non-collaborative settings, demonstrate the effectiveness of UDP in learning user-specific dialogue strategies. Results validate the protocol's utility and highlight UDP's robustness, adaptability, and potential to advance user-centric dialogue systems.