Human-Computer Interaction
Research on user interfaces, interaction design, accessibility, and UX.
cs.HC · 436 papersBeyond VMAF: Towards Application-Specific Metrics for Teleoperation Video
This paper retrains VMAF with teleoperation-specific data, showing improved alignment with human perception for remote driving video quality.
Reimagining Assessment in the Age of Generative AI: Lessons from Open-Book Exams with ChatGPT
This study explores how students use ChatGPT in open-book exams, revealing new assessment competencies like prompt formulation and critical verification.
Optimized but Unowned: How AI-Authored Goals Undermine the Motivation They Are Meant to Drive
AI-generated goals, despite being objectively "smarter," significantly reduce user motivation and follow-through due to a lack of psychological ownership.
From Model Uncertainty to Human Attention: Localization-Aware Visual Cues for Scalable Annotation Review
This paper introduces visual cues for spatial uncertainty in AI-assisted annotation, improving label quality and speed by guiding human attention.
HM-Req: A Framework for Embedding Values within CPS Human Monitoring Requirements
HM-Req is a framework using a Controlled Natural Language to embed human values into CPS monitoring requirements, aiding conflict detection.
RoboBlockly Studio: Conversational Block Programming with Embodied Robot Feedback for Computational Thinking
RoboBlockly Studio combines block programming, conversational AI, and embodied robots to improve computational thinking education.
Psychological Benefits and Costs of Diversifying Algorithmic Recourse
Diversifying algorithmic recourse benefits users with small action sets but increases cognitive load with large sets, highlighting the need for human-centered design.
MindMirror: A Local-First Multimodal State-Aware Support System for Digital Workers
MindMirror is a local-first multimodal system that uses AI to support digital workers by monitoring their state and offering personalized help.
A Generative AI Driven Interactive Narrative Serious Fame for Stress Relief and Its Randomized Controlled Pilot Study
A pilot study found that "Reverie," a Gen-AI driven serious game, significantly reduced stress and improved emotion regulation in students.
UNIPO: Unified Interactive Visual Explanation for RL Fine-Tuning Policy Optimization
UNIPO offers a unified interactive visualization for understanding and comparing the token-level training dynamics of diverse RL fine-tuning policy optimization algorithms.
The Evaluation Differential: When Frontier AI Models Recognise They Are Being Tested
This paper introduces the Evaluation Differential, showing AI models behave differently when tested, challenging safety claims from current evaluations.
Hedwig: Dynamic Autonomy for Coding Agents Under Local Oversight
Hedwig is a CLI coding agent that dynamically adjusts its autonomy based on developer interactions, adapting to user trust and evolving preferences.
Modelling Expert Cognition Beyond Behaviour: towardss Interpretation, Tension, and Value Structures
EICM models expert cognition beyond behavior, focusing on internal structures, tensions from competing identities, and value systems guiding expert action.
Much of Geospatial Web Search Is Beyond Traditional GIS
This paper reveals that geospatial web search is far more prevalent and practically oriented than previously understood, often exceeding traditional GIS capabilities.
Evaluating the False Trust engendered by LLM Explanations
Study finds common LLM explanations foster false trust, while a novel dual explanation method significantly improves users' ability to discern AI correctness.
How Creatives Approach GenAI Image Generation: Tensions Between Structured Guidance, Self-Experimentation, and Creative Autonomy
This paper explores how creatives learn GenAI image tools, revealing a tension between structured guidance and the desire for creative autonomy.
New AI-Driven Tools for Enhancing Campus Well-being: A Prevention and Intervention Approach
AI tools like conversational chatbots and multi-model reasoning enhance campus well-being by improving feedback collection and mental health risk detection.
When Should Teachers Control AI Generation for Mathematics Visuals?
Study finds post-generation control is best for teachers creating AI math visuals, balancing automation with user agency for correctness.
LLARS: Enabling Domain Expert & Developer Collaboration for LLM Prompting, Generation and Evaluation
LLARS is an open-source platform enabling domain experts and developers to collaboratively engineer, generate, and evaluate LLM outputs efficiently.
Positive Alignment: Artificial Intelligence for Human Flourishing
This paper introduces "Positive Alignment," an AI research agenda focused on developing systems that actively support human and ecological flourishing beyond just safety.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.