help研究问题人机协同标注与示例选择算法学习中,如何通过参与者协作生成高质量的子目标标签以替代专家创建?CHI '22AlgoSolve: Supporting Subgoal Learning in Algorithmic Problem-Solving with Learnersourced Microtasks
help研究问题人机协同标注与示例选择两阶段子目标微任务设计(投票与标注)如何影响子目标标签的质量和学习效果?CHI '22AlgoSolve: Supporting Subgoal Learning in Algorithmic Problem-Solving with Learnersourced Microtasks
help研究问题人机协同标注与示例选择在算法问题解决中,结合多臂赌博机算法是否能动态优化子目标学习任务?CHI '22AlgoSolve: Supporting Subgoal Learning in Algorithmic Problem-Solving with Learnersourced Microtasks
lightbulb现实痛点人机协同标注与示例选择初学者在算法学习中难以规划完整的解决方案。CHI '22AlgoSolve: Supporting Subgoal Learning in Algorithmic Problem-Solving with Learnersourced Microtasks
help研究问题人机协同标注与示例选择专家的快速决策规则(如布尔公式)如何用于改进机器学习模型的领域适应性?IUI '21Decision rule elicitation for domain adaptation
help研究问题人机协同标注与示例选择将专家规则与机器学习模型整合时,如何处理规则冲突并优化模型性能?IUI '21Decision rule elicitation for domain adaptation
help研究问题人机协同标注与示例选择专家生成规则的复杂性和一致性如何影响模型的跨域泛化能力?IUI '21Decision rule elicitation for domain adaptation
lightbulb现实痛点人机协同标注与示例选择AI模型在跨领域测试时表现差且需大量标注数据支持。IUI '21Decision rule elicitation for domain adaptation
help研究问题人机协同标注与示例选择复杂环境背景(道路周围建筑密度等)如何影响高度自动驾驶中的控制权转移时间?IUI '21The Effect of Surrounding Scenery Complexity on the Transfer of Control Time in Highly Automated Driving
help研究问题人机协同标注与示例选择在不同环境复杂度下,次要任务的存在如何改变控制权转移时间?IUI '21The Effect of Surrounding Scenery Complexity on the Transfer of Control Time in Highly Automated Driving
lightbulb现实痛点人机协同标注与示例选择在自动驾驶中,司机在复杂环境下接管车辆容易延误,存在安全隐患IUI '21The Effect of Surrounding Scenery Complexity on the Transfer of Control Time in Highly Automated Driving
help研究问题人机协同标注与示例选择AI辅助系统如何提升多标签数据标注任务的准确性与效率?IUI '21Increasing the Speed and Accuracy of Data Labeling Through an AI Assisted Interface
help研究问题人机协同标注与示例选择如何利用计算机视觉技术将图标图像自动转换为矢量字体以提高加载速度?IUI '21Auto-Icon: An Automated Code Generation Tool for Icon Designs Assisting in UI Development
help研究问题人机协同标注与示例选择深度学习模型在图标分类标签的预测中如何减少开发者的标注工作量?IUI '21Auto-Icon: An Automated Code Generation Tool for Icon Designs Assisting in UI Development
help研究问题人机协同标注与示例选择在UI开发中,如何通过颜色属性检测增强代码的描述性?IUI '21Auto-Icon: An Automated Code Generation Tool for Icon Designs Assisting in UI Development
lightbulb现实痛点人机协同标注与示例选择图标设计耗时且代码难以维护,影响UI开发效率。IUI '21Auto-Icon: An Automated Code Generation Tool for Icon Designs Assisting in UI Development
help研究问题人机协同标注与示例选择如何通过结合主动学习和游戏化提升众包数据注释的质量?UbiComp '21CrowdAct: Achieving High-Quality Crowdsourced Datasets in Mobile Activity Recognition
help研究问题人机协同标注与示例选择在活动识别数据注释中,不准确数据检测算法如何影响整体质量?UbiComp '21CrowdAct: Achieving High-Quality Crowdsourced Datasets in Mobile Activity Recognition
help研究问题人机协同标注与示例选择游戏化设计在众包活动数据收集中的实际效果如何?UbiComp '21CrowdAct: Achieving High-Quality Crowdsourced Datasets in Mobile Activity Recognition
lightbulb现实痛点人机协同标注与示例选择众包数据标注准确性差、激励不足,导致活动识别模型难以可靠应用。UbiComp '21CrowdAct: Achieving High-Quality Crowdsourced Datasets in Mobile Activity Recognition
help研究问题人机协同标注与示例选择在各类仓库环境中,如何利用物理信息生成更接近真实场景的合成数据?UbiComp '21PIWIMS: Physics Informed Warehouse Inventory Monitoring via Synthetic Data Generation
help研究问题人机协同标注与示例选择物理信息指导下的合成数据能否有效减少手动标注需求并提升库存监控精度?UbiComp '21PIWIMS: Physics Informed Warehouse Inventory Monitoring via Synthetic Data Generation
help研究问题人机协同标注与示例选择PIWIMS方法在不同相机配置和产品移动场景下的泛化性能如何表现?UbiComp '21PIWIMS: Physics Informed Warehouse Inventory Monitoring via Synthetic Data Generation
lightbulb现实痛点人机协同标注与示例选择现有基于摄像头的库存监控系统在不同仓库环境中泛化性差,标注成本高。UbiComp '21PIWIMS: Physics Informed Warehouse Inventory Monitoring via Synthetic Data Generation
help研究问题人机协同标注与示例选择如何利用天花板摄像机图片估算用户垂直视野中的亮度分布?UbiComp '21LumNet: Learning to Estimate Vertical Visual Field Luminance for Adaptive Lighting Control
help研究问题人机协同标注与示例选择深度学习模型能否替代现有基于线性模型的亮度估算方法?UbiComp '21LumNet: Learning to Estimate Vertical Visual Field Luminance for Adaptive Lighting Control
help研究问题人机协同标注与示例选择自监督学习是否能够在少量标注数据下提高亮度估算的准确性?UbiComp '21LumNet: Learning to Estimate Vertical Visual Field Luminance for Adaptive Lighting Control
lightbulb现实痛点人机协同标注与示例选择现有亮度控制系统依赖侵入式设备,妨碍用户使用。UbiComp '21LumNet: Learning to Estimate Vertical Visual Field Luminance for Adaptive Lighting Control
help研究问题人机协同标注与示例选择如何减少在球类运动视频事件标注中的人工交互?CHI '21EventAnchor: Reducing Human Interactions in Event Annotation of Racket Sports Videos
help研究问题人机协同标注与示例选择在低质量(如转播)视频中,现有计算机视觉系统如何改进以更准确地检测高层次事件信息?CHI '21EventAnchor: Reducing Human Interactions in Event Annotation of Racket Sports Videos
help研究问题人机协同标注与示例选择如何设计一个可扩展且高效的框架,用于快节奏运动的事件标注?CHI '21EventAnchor: Reducing Human Interactions in Event Annotation of Racket Sports Videos
lightbulb现实痛点人机协同标注与示例选择分析乒乓球赛视频耗时且现有工具难满足战术分析需求。CHI '21EventAnchor: Reducing Human Interactions in Event Annotation of Racket Sports Videos
help研究问题人机协同标注与示例选择空间布局设计如何帮助非专家提高图像标注质量?CHI '21Spatial Labeling: Leveraging Spatial Layout for Improving Label Quality in Non-Expert Image Annotation
help研究问题人机协同标注与示例选择在复杂和模糊图像标注任务中,空间布局的效果如何?CHI '21Spatial Labeling: Leveraging Spatial Layout for Improving Label Quality in Non-Expert Image Annotation
lightbulb现实痛点人机协同标注与示例选择非专家标注图像错误率高,影响机器学习模型性能。CHI '21Spatial Labeling: Leveraging Spatial Layout for Improving Label Quality in Non-Expert Image Annotation
help研究问题人机协同标注与示例选择社会生态系统中,哪些因素影响无陪伴移民青少年使用心理健康App?CHI '21Unaccompanied Migrant Youth and Mental Health Technologies: A Social-Ecological Approach to Understanding and Designing
help研究问题人机协同标注与示例选择现有心理健康App如何无法满足无陪伴移民青少年的特殊需求?CHI '21Unaccompanied Migrant Youth and Mental Health Technologies: A Social-Ecological Approach to Understanding and Designing
help研究问题人机协同标注与示例选择数据级联效应(由数据质量问题引发的连锁反应)的触发因素和特征是什么?CHI '21"Everyone wants to do the model work, not the data work": Data Cascades in High-Stakes AI
help研究问题人机协同标注与示例选择高风险领域(如医疗、环境保护)中,如何通过跨组织合作和标准化文档减少数据级联效应的发生?CHI '21"Everyone wants to do the model work, not the data work": Data Cascades in High-Stakes AI
help研究问题人机协同标注与示例选择在人工智能开发中,如何定义和衡量数据质量指标以更好地应对数据问题?CHI '21"Everyone wants to do the model work, not the data work": Data Cascades in High-Stakes AI
lightbulb现实痛点人机协同标注与示例选择高风险领域的AI难以规避因数据质量问题导致的错误预测。CHI '21"Everyone wants to do the model work, not the data work": Data Cascades in High-Stakes AI
help研究问题人机协同标注与示例选择在包含人类意见分歧的数据中,如何有效评估机器学习模型的实际性能?CHI '21The Disagreement Deconvolution: Bringing Machine Learning Performance Metrics In Line With Reality
help研究问题人机协同标注与示例选择基于多重标注数据,如何减少模型评价中的噪音和主观分歧?CHI '21The Disagreement Deconvolution: Bringing Machine Learning Performance Metrics In Line With Reality
help研究问题人机协同标注与示例选择现有机器学习性能指标是否能准确反映模型在实际社交计算任务中的表现?CHI '21The Disagreement Deconvolution: Bringing Machine Learning Performance Metrics In Line With Reality
lightbulb现实痛点人机协同标注与示例选择社交计算任务中,机器学习模型常因忽略人类分歧导致表现评估过于乐观。CHI '21The Disagreement Deconvolution: Bringing Machine Learning Performance Metrics In Line With Reality
lightbulb现实痛点人机协同标注与示例选择领域专家标注数据的过程复杂且耗时,标签质量无法保障。CHI '21Designing Ground Truth and the Social Life of Labels