首页/论文/2001.02271

arXiv 2001.022712020-01-09cs.HCcs.AI

Revealing Neural Network Bias to Non-Experts Through Interactive Counterfactual Examples

Drexel University（德雷塞尔大学）；IT University of Copenhagen（哥本哈根信息技术大学）

Chelsea M. Myers, Evan Freed, Luis Fernando Laris Pardo, Anushay Furqan, Sebastian Risi, Jichen Zhu

英文摘要：

AI algorithms are not immune to biases. Traditionally, non-experts have little control in uncovering potential social bias (e.g., gender bias) in the algorithms that may impact their lives. We present a preliminary design for an interactive visualization tool CEB to reveal biases in a commonly used AI method, Neural Networks (NN). CEB combines counterfactual examples and abstraction of an NN decision process to empower non-experts to detect bias. This paper presents the design of CEB and initial findings of an expert panel (n=6) with AI, HCI, and Social science experts.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

阅读原文

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

本文内容

英文摘要相关论文

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

继续探索

相关论文

深度学习交通预测的反事实解释

Counterfactual Explanations for Deep Learning-Based Traffic Forecasting

ETH Zurich（苏黎世联邦理工学院）；WSL Institute for Snow and Avalanche Research SLF（WSL雪与雪崩研究所SLF）

本研究提出了一种基于反事实解释的框架，用于提升深度学习交通预测模型的可解释性，通过场景驱动的约束生成定制化解释，以揭示模型学习的交通模式。

2024-05-01cs.AIcs.LG

忠实、充分且可解释：通过离散扩散反演重新思考图反事实解释

Faithful, Sufficient and Understandable: Rethinking Graph Counterfactual Explanations via Discrete Diffusion Inversion

TU Berlin（柏林工业大学）；BIFOLD–Berlin Institute for the Foundations of Learning and Data（柏林学习与数据基础研究所）

该研究针对图神经网络预测缺乏内在解释的问题，提出GDCE-I方法，结合离散扩散反演技术，在统一评估框架下于四个基准测试中大幅优于相关工作，可生成符合领域规则的可解释图反事实解释。

2026-08-13cs.AIcs.LG

Interactive Classification for Deep Learning Interpretation

Georgia Institute of Technology（佐治亚理工学院）

We present an interactive system enabling users to manipulate images to explore the robustness and sensitivity of deep learning image classifiers. Using modern web technologies to run in-browser inference, users can remove image features using inpainting algorithms and obtain new classifications in real time, which allows them to ask a variety of "what if" questions by experimentally modifying images and seeing how the model reacts. Our system allows users to compare and contrast what image regions humans and machine learning models use for classification, revealing a wide range of surprising …

2019-04-10cs.AIcs.CVcs.HC

理解大型语言模型行为：通过交互式反事实生成与分析

Understanding Large Language Model Behaviors through Interactive Counterfactual Generation and Analysis

ETH Zürich（苏黎世联邦理工学院）；University of Konstanz（康斯坦茨大学）；IBM Research（IBM研究院）

本文提出LLM Analyzer交互式可视化系统，通过反事实分析高效探索LLM行为，生成语义反事实并计算特征归因，经用户研究验证其有效性与人机协作价值。

2025-08-07cs.AIcs.CLcs.HC

交互式单变量时间序列反事实生成

Interactive Counterfactual Generation for Univariate Time Series

University of Konstanz（康斯坦茨大学）

提出一种交互式方法，利用二维投影和决策边界图生成单变量时间序列分类的反事实解释，通过ECG5000数据集验证，显著提升可解释性和用户理解。

2024-08-20cs.HCcs.LG

更多相关论文收起更多论文

VISION：利用反事实增强的鲁棒且可解释的代码漏洞检测

VISION: Robust and Interpretable Code Vulnerability Detection Leveraging Counterfactual Augmentation

VISION 框架利用 LLM 生成反事实样本增强训练数据，结合 GNN 和可解释性分析，显著提升代码漏洞检测的鲁棒性和准确性，并发布 CWE-20-CFA 基准。

2025-08-26cs.AIcs.CRcs.CY

哪些虚假相关性影响NLI模型的推理？通过数据约束反事实的视觉交互式诊断

Which Spurious Correlations Impact Reasoning in NLI Models? A Visual Interactive Diagnosis through Data-Constrained Counterfactuals

ETH Zürich（苏黎世联邦理工学院）；ETH AI Center（苏黎世联邦理工学院人工智能中心）

本文提出一种人机交互仪表盘，利用GPT-3生成反事实示例并获取模型反馈，以诊断NLI模型依赖的虚假相关性，将其归为语义相关性、逻辑谬误和偏见三类，并指出多样化训练数据与构建对抗性测试套件等研究机会。

2023-06-21cs.CLcs.HC

ViCE: Visual Counterfactual Explanations for Machine Learning Models

New York University Abu Dhabi（纽约大学阿布扎比分校）；NYU Tandon School of Engineering（纽约大学坦登工程学院）

The continued improvements in the predictive accuracy of machine learning models have allowed for their widespread practical application. Yet, many decisions made with seemingly accurate models still require verification by domain experts. In addition, end-users of a model also want to understand the reasons behind specific decisions. Thus, the need for interpretability is increasingly paramount. In this paper we present an interactive visual analytics tool, ViCE, that generates counterfactual explanations to contextualize and evaluate model decisions. Each sample is assessed to identify the m…

2020-03-05cs.AIcs.HCcs.LG

揭示未写之物：束搜索树的可视化调查以应对语言模型提示挑战

Revealing the Unwritten: Visual Investigation of Beam Search Trees to Address Language Model Prompting Challenges

University of Konstanz（康斯坦茨大学）；ETH Zürich（苏黎世联邦理工学院）

针对大型语言模型提示挑战，提出交互式可视化束搜索树的方法，以分析模型生成决策，验证结果并提供新见解。

2023-10-17cs.AIcs.CLcs.HC

TofuML：一款面向新手的机器学习交互式探索用空间物理交互机器学习设备

TofuML: A Spatio-Physical Interactive Machine Learning Device for Interactive Exploration of Machine Learning for Novices

The University of Tokyo（东京大学）；Tsukuba University（筑波大学）；Fujitsu Limited（富士通株式会社）

本研究推出面向非专业用户的物理空间交互ML系统TofuML，通过两项用户研究证实其相比GUI版本更能提升用户参与度、降低ML入门门槛，为普适性交互式ML系统开发提供依据。

2025-08-06cs.HC

MiMICRI：迈向以领域为中心的心血管图像分类模型反事实解释

MiMICRI: Towards Domain-centered Counterfactual Explanations of Cardiovascular Image Classification Models

Georgia Institute of Technology（佐治亚理工学院）；Cleveland Clinic（克利夫兰诊所）；IBM Research（IBM研究院）

提出框架与Python库MiMICRI，通过交互式替换医学图像形态学结构生成反事实解释，帮助专家评估模型预测影响并验证医学事实，提升了心血管图像分类模型在医疗领域的可解释性与可信度。

2024-04-24cs.CVcs.HCcs.LG

Segue: Overviewing Evolution Patterns of Egocentric Networks by Interactive Construction of Spatial Layouts

Getting the overall picture of how a large number of ego-networks evolve is a common yet challenging task. Existing techniques often require analysts to inspect the evolution patterns of ego-networks one after another. In this study, we explore an approach that allows analysts to interactively create spatial layouts in which each dot is a dynamic ego-network. These spatial layouts provide overviews of the evolution patterns of ego-networks, thereby revealing different global patterns such as trends, clusters and outliers in evolution patterns. To let analysts interactively construct interpreta…

2018-11-12cs.HCcs.SI

LangLasso：通过LLM解释实现交互式聚类描述

LangLasso: Interactive Cluster Descriptions through LLM Explanation

University of Konstanz（康斯坦茨大学）；Utrecht University（乌得勒支大学）；Eindhoven University of Technology（埃因霍温理工大学）

LangLasso通过LLM生成交互式自然语言描述，使非专家也能理解聚类结构并整合外部知识。

2026-01-15cs.HCcs.LGstat.CO

Embodiment in Socially Interactive Robots

University of Southern California（南加州大学）；University of Wisconsin–Madison（威斯康星大学麦迪逊分校）

Physical embodiment is a required component for robots that are structurally coupled with their real-world environments. However, most socially interactive robots do not need to physically interact with their environments in order to perform their tasks. When and why should embodied robots be used instead of simpler and cheaper virtual agents? This paper reviews the existing work that explores the role of physical embodiment in socially interactive robots. This class consists of robots that are not only capable of engaging in social interaction with humans, but are using primarily their social…

2019-12-01cs.HCcs.RO

Regshock: Interactive Visual Analytics of Systemic Risk in Financial Networks

Financial regulatory agencies are struggling to manage the systemic risks attributed to negative economic shocks. Preventive interventions are prominent to eliminate the risks and help to build a more resilient financial system. Although tremendous efforts have been made to measure multi-risk severity levels, understand the contagion behaviors and other risk management problems, there still lacks a theoretical framework revealing what and how regulatory intervention measurements can mitigate systemic risk. Here we demonstrate regshock, a practical visual analytical approach to support the expl…

2021-04-24cs.HCcs.SIq-fin.RM

基于交互式大语言模型的痴呆相关日常活动模拟器

An Interactive LLM-Based Simulator for Dementia-Related Activities of Daily Living

Worcester Polytechnic Institute（伍斯特理工学院）；University of California Davis（加州大学戴维斯分校）；MGH Institute of Health Professionals（麻省总医院健康专业学院）；University of New Hampshire（新罕布什尔大学）

本文提出一个交互式模拟器，利用大语言模型生成痴呆患者在日常活动中的多轮行为，通过专家反馈优化策略，提升辅助AI和机器人在痴呆护理中的应用效果。

2026-03-06cs.GRcs.HCcs.RO