检索结果-内蒙古大学图书馆

Robustness Evaluation of the German Extractive Question Answering Task 31

学校读者我要写书评

暂无评论

Robustness Evaluation of the German Extractive Question Answ...

31st International Conference on Computational Linguistics, COLING 2025

作者： Satheesh, Shalaka Beckh, Katharina Klug, Katrin Allende-Cid, Héctor Houben, Sebastian Hassan, Teena Fraunhofer Institute for Intelligent Analysis and Information Systems IAIS Germany Lamarr Institute for Machine Learning and Artificial Intelligence Bonn-Rhein-Sieg University of Applied Sciences Germany

ISBN: (纸本)9798891761964

To ensure reliable performance of Question Answering (QA) systems, evaluation of robustness is crucial. Common evaluation benchmarks commonly only include performance metrics, such as Exact Match (EM) and the F1 score. However, these benchmarks overlook critical factors for the deployment of QA systems. This oversight can result in systems vulnerable to minor perturbations in the input such as typographical errors. While several methods have been proposed to test the robustness of QA models, there has been minimal exploration of these approaches for languages other than English. This study focuses on the robustness evaluation of German language QA models, extending methodologies previously applied primarily to English. The objective is to nurture the development of robust models by defining an evaluation method specifically tailored to the German language. We assess the applicability of perturbations used in English QA models for German and perform a comprehensive experimental evaluation with eight models. The results show that all models are vulnerable to character-level perturbations. Additionally, the comparison of monolingual and multilingual models suggest that the former are less affected by character and word-level perturbations. © 2025 Association for Computational Linguistics.

关键词： Computational linguistics

RoboCup@Home 2024 OPL Winner NimbRo: Anthropomorphic Service Robots Using Foundation Models for Perception and Planning

学校读者我要写书评

暂无评论

RoboCup@Home 2024 OPL Winner NimbRo: Anthropomorphic Service...

27th RoboCup International Symposium, 2024

作者： Memmesheimer, Raphael Nogga, Jan Pätzold, Bastian Kruzhkov, Evgenii Bultmann, Simon Schreiber, Michael Bode, Jonas Karacora, Bertan Park, Juhui Savinykh, Alena Behnke, Sven Autonomous Intelligent Systems Computer Science Institute VI Lamarr Institute for Machine Learning and Artificial Intelligence and Center for Robotics University of Bonn Bonn Germany

ISBN: (纸本)9783031858581

We present the approaches and contributions of the winning team NimbRo@Home at the RoboCup@Home 2024 competition in the Open Platform League held in Eindhoven, NL. Further, we describe our hardware setup and give an overview of the results for the task stages and the final demonstration. For this year’s competition, we put a special emphasis on open-vocabulary object segmentation and grasping approaches that overcome the labeling overhead of supervised vision approaches, commonly used in RoboCup@Home. We successfully demonstrated that we can segment and grasp non-labeled objects by text descriptions. Further, we extensively employed Large Language Model (LLMs) for natural language understanding and task planning. Throughout the competition, our approaches showed robustness and generalization capabilities. A video of our performance can be found online (https://***/videos/RoboCup_2024). © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Robot programming

Object-Centric Image to Video Generation with Language Guidance

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Villar-Corrales, Angel Plepi, Gjergj Behnke, Sven Autonomous Intelligent Systems Computer Science Institute VI – Intelligent Systems and Robotics Center for Robotics Lamarr Institute for Machine Learning and Artificial Intelligence Germany

Accurate and flexible world models are crucial for autonomous systems to understand their environment and predict future events. Object-centric models, with structured latent spaces, have shown promise in modeling object dynamics and interactions, but often face challenges in scaling to complex datasets and incorporating external guidance, limiting their applicability in robotics. To address these limitations, we propose TextOCVP, an object-centric model for image-to-video generation guided by textual descriptions. TextOCVP parses an observed scene into object representations, called slots, and utilizes a text-conditioned transformer predictor to forecast future object states and video frames. Our approach jointly models object dynamics and interactions while incorporating textual guidance, thus leading to accurate and controllable predictions. Our method’s structured latent space offers enhanced control over the prediction process, outperforming several image-to-video generative baselines. Additionally, we demonstrate that structured object-centric representations provide superior controllability and interpretability, facilitating the modeling of object dynamics and enabling more precise and understandable predictions. Videos and code are available at https://***/TextOCVP/. © 2025, CC BY.

关键词： Digital elevation model

Towards Conscious Service Robots

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Behnke, Sven Autonomous Intelligent Systems Computer Science Institute VI – Intelligent Systems and Robotics Center for Robotics the Lamarr Institute for Machine Learning and Artificial Intelligence University of Bonn Germany

Deep learning’s success in perception, natural language processing, etc. inspires hopes for advancements in autonomous robotics. However, real-world robotics face challenges like variability, high-dimensional state spaces, non-linear dependencies, and partial observability. A key issue is non-stationarity of robots, environments, and tasks, leading to performance drops with out-of-distribution data. Unlike current machine learning models, humans adapt quickly to changes and new tasks due to a cognitive architecture that enables systematic generalization and meta-cognition. Human brain’s System 1 handles routine tasks unconsciously, while System 2 manages complex tasks consciously, facilitating flexible problem-solving and self-monitoring. For robots to achieve human-like learning and reasoning, they need to integrate causal models, working memory, planning, and metacognitive processing. By incorporating human cognition insights, the next generation of service robots will handle novel situations and monitor themselves to avoid risks and mitigate errors. Copyright © 2025, The Authors. All rights reserved.

关键词： Problem solving

Person Segmentation and Action Classification for Multi-Channel Hemisphere Field of View LiDAR Sensors

学校读者我要写书评

暂无评论

Person Segmentation and Action Classification for Multi-Chan...

IEEE/SICE International Symposium on System Integration

作者： Svetlana Seliunina Artem Otelepko Raphael Memmesheimer Sven Behnke Autonomous Intelligent Systems Group Computer Science Institute VI – Intelligent Systems and Robotics Lamarr Institute for Machine Learning and Artificial Intelligence and Center for Robotics University of Bonn

ISBN: (数字)9798331531614

ISBN: (纸本)9798331531621

关键词： Reflectivity Adaptation models Laser radar Three-dimensional displays System integration Robot sensing systems Encoding Sensor systems Sensors Safety

PlaySlot: learning Inverse Latent Dynamics for Controllable Object-Centric Video Prediction and Planning

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Villar-Corrales, Angel Behnke, Sven Computer Science Institute VI – Intelligent Systems and Robotics Center for Robotics The Lamarr Institute for Machine Learning and Artificial Intelligence Germany

Predicting future scene representations is a crucial task for enabling robots to understand and interact with the environment. However, most existing methods rely on video sequences and simulations with precise action annotations, limiting their ability to leverage the large amount of available unlabeled video data. To address this challenge, we propose PlaySlot, an object-centric video prediction model that infers object representations and latent actions from unlabeled video sequences. It then uses these representations to forecast future object states and video frames. PlaySlot allows to generate multiple possible futures conditioned on latent actions, which can be inferred from video dynamics, provided by a user, or generated by a learned action policy, thus enabling versatile and interpretable world modeling. Our results show that PlaySlot outperforms both stochastic and object-centric baselines for video prediction across different environments. Furthermore, we show that our inferred latent actions can be used to learn robot behaviors sample-efficiently from unlabeled video demonstrations. Videos and code are available at https://***/PlaySlot/. © 2025, CC BY.

关键词： Stochastic systems

LiLMaps: Learnable Implicit Language Maps

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Kruzhkov, Evgenii Behnke, Sven Autonomous Intelligent Systems Computer Science Institute VI University of Bonn Germany Autonomous Intelligent Systems Computer Science Institute VI – Intelligent Systems and Robotics Center for Robotics and The Lamarr Institute for Machine Learning and Artificial Intelligence University of Bonn Germany

One of the current trends in robotics is to employ large language models (LLMs) to provide non-predefined command execution and natural human-robot interaction. It is useful to have an environment map together with its language representation, which can be further utilized by LLMs. Such a comprehensive scene representation enables numerous ways of interaction with the map for autonomously operating robots. In this work, we present an approach that enhances incremental implicit mapping through the integration of vision-language features. Specifically, we (i) propose a decoder optimization technique for implicit language maps which can be used when new objects appear on the scene, and (ii) address the problem of inconsistent vision-language predictions between different viewing positions. Our experiments demonstrate the effectiveness of LiLMaps and solid improvements in performance. © 2025, CC BY-NC-ND.

关键词： Visual languages

Adaptive deep probabilistic regression for real-time motor excitability state prediction from human EEG

学校读者我要写书评

暂无评论

Brain Stimulation 2025年第1期18卷 400-401页

作者： Haxel, Lisa Kapoor, Jaivardhan Ahola, Oskari Kahilakoski, Olli-Pekka Kirchhoff, Miriam Roine, Timo Ziemann, Ulf Macke, Jakob H. Machine Learning in Science Excellence Cluster Machine Learning & Tübingen AI Center Germany Hertie Institute for Clinical Brain Research Department Neurology and Stroke Germany Department of Neuroscience and Biomedical Engineering Aalto University School of Science Finland Department Empirical Inference Max Planck Institute for Intelligent Systems Germany

LoGAN: A novel local attentive generative adversarial resizable network for detailed 3D reconstruction of the Martian surface using monocular HiRISE images and DTMs

学校读者我要写书评

暂无评论

ISPRS Journal of Photogrammetry and Remote Sensing 2025年 225卷 302-327页

作者： Cao, Jiarui Huang, Rong Xu, Yusheng Ye, Zhen Qian, Jia Hoegner, Ludwig Tong, Xiaohua College of Surveying and Geoinformatics Tongji University Shanghai China Shanghai Key Laboratory for Planetary Mapping and Remote Sensing for Deep Space Exploration Shanghai China Institute for Applications of Machine Learning and Intelligent Systems Munich University of Applied Sciences Munich Germany

High-resolution Mars digital terrain models (DTMs) are crucial for exploration missions and scientific research. However, traditional photogrammetry is constrained by stereo image coverage and configuration, while laser altimetry suffers from sparse laser footprint distribution, leading to limited terrain accuracy. While the recent photoclinometric and deep learning methods can generate promising 3D reconstruction results, they still face challenges such as complex parameterization. To address these drawbacks, we present a novel attentive generative adversarial resizable network for Martian terrain reconstruction, which can generate 3D terrain models with rich details from high-resolution monocular images. In this proposed method, a novel local attention module is embedded to investigate geometric dependencies further, preserving spatial correlations in local contexts. A new combination of loss functions is adopted to guide the generation process, simultaneously suppressing artifacts and recovering details. A multi-scale alternate training strategy is also introduced, which performs well in preserving details at various scales. Experimental results in areas of Hephaestus Fossae, Oxia Palus, and Victoria Crater demonstrate that our presented approach can generate large-scale DTMs with a spatial resolution of pixel level (0.25 m/pixel), showing more details of textures and equivalent elevation accuracy. The reconstruction accuracy is similar to the most accurate photoclinometric results but without the need for complex modeling. © 2025 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS)

关键词： Martian surface analysis