Trust in Vision-Language Models: Insights from a Participatory User Workshop

2025-11-17

AI Summary

To address the lack of clarity regarding user trust mechanisms in vision-language models (VLMs), this study pioneers a participatory workshop methodology, systematically collecting contextualized trust perception data through iterative, real-user interactions and co-design activities. Departing from conventional AI-performance–centric evaluation paradigms, it adopts an empirically grounded, user-centered perspective to develop a trust assessment framework structured along four dimensions: task context, explainability, controllability, and error recovery. The research identifies key dynamic factors shaping trust evolution—particularly explanation consistency and feedback timeliness—and proposes a scalable, qualitative–design–integrated methodology. This work provides both theoretical foundations and actionable design guidelines for trust-oriented VLM interfaces, advancing trustworthy AI research from a technology-centric to a human-centered paradigm.

Technology Category

• Computer Vision: Language and Vision
• Philosophy and Ethics of AI: Safety, Robustness & Trustworthiness
• Natural Language Processing: Safety and Robustness

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

With the growing deployment of Vision-Language Models (VLMs), pre-trained on large image-text and video-text datasets, it is critical to equip users with the tools to discern when to trust these systems. However, examining how user trust in VLMs builds and evolves remains an open problem. This problem is exacerbated by the increasing reliance on AI models as judges for experimental validation, to bypass the cost and implications of running participatory design studies directly with users. Following a user-centred approach, this paper presents preliminary results from a workshop with prospective VLM users. Insights from this pilot workshop inform future studies aimed at contextualising trust metrics and strategies for participants' engagement to fit the case of user-VLM interaction.

Problem

• Understanding how user trust in Vision-Language Models builds and evolves
• Addressing reliance on AI models instead of participatory user studies
• Developing contextualized trust metrics for user-VLM interaction strategies

Innovation

• User-centered participatory workshop for trust evaluation
• Contextualizing trust metrics for VLM interaction
• Engaging participants to inform future trust studies