VAxAutoSci: Visual Analytics in the era of Autonomous Science

VISUAL ANALYTICS

in the era of

AUTONOMOUS SCIENCE

IEEE VIS 2026 WORKSHOP

Boston · November 9

Workshop Overview

Artificial intelligence is rapidly transforming scientific workflows across domains. For example, in emerging autonomous laboratories (e.g., self-driving labs that autonomously synthesize and test new materials), artificial agents design experiments, analyze results, and iteratively refine hypotheses within closed-loop pipelines, thus fundamentally transforming the role of scientists.

Such a transition creates new requirements and opportunities. Visual analytics enables oversight and steering of autonomous processes, facilitates the inspection and refinement of machine-generated hypotheses, and supports effective human-AI collaboration in knowledge generation and scientific discovery. This workshop positions visual analytics as a core enabler of autonomous scientific discovery and advances two complementary directions:

Developing interactive visual interfaces that support AI-accelerated science, and
Leveraging AI-accelerated scientific platforms to advance visualization research into AI-driven workflows.

We encourage submissions at the intersection of visual analytics, AI-accelerated scientific discovery, and various application domains (e.g., materials science). Submissions may address important questions including, but not limited to, analyzing autonomous decisions by AI or understanding how humans should intervene in closed-loop pipelines.

Goals

Our workshop aims to accomplish the following:

Identify research opportunities within VIS for autonomous laboratories
Share early-stage research, systems, applications, workflows, and ideas
Build a cross-disciplinary autonomous labs research community centered around VIS
Foster new connections between researchers and scientific domains

Topics

Our topics of interest include, but are not limited to, the following:

Visualization, visual analytics, or immersive analytics for autonomous scientific discovery
Evaluation of methods for designing, monitoring, and steering autonomous scientific processes
Guidelines and strategies for human-AI collaborative scientific discovery and knowledge generation from multi-modal data
Ethical and safety considerations of autonomous science and the role of visual analytics in mitigating risks
Case studies from application domains (e.g., physics, chemistry, materials science, drug discovery, biology, etc.)
Use cases of leveraging AI to accelerate the science of visualization and analytic reasoning (e.g., adaptive user studies)
The hand-off between human scientists and AI scientists in closed-loop workflows

Call for Participation

Short Papers

We invite 4–6 page manuscripts (excluding references). Accepted papers will be published in IEEE Xplore with authors’ permission. Submissions will be peer-reviewed by one member of the organizing committee and two members of the program committee. We will ensure that reviewers do not have any conflicts of interest with the authors. Submissions will be evaluated based on (1) alignment with the workshop theme, (2) technical soundness and rigor, and (3) potential to facilitate engaging discussions.

Late Breaking Work

We invite 2–3 page manuscripts (excluding references) on the latest tools, findings, workflows, and viewpoints at the intersection of VIS and autonomous science. Submissions are non-archival and will only be accessible from the workshop website. Submissions will be reviewed by one member of the organizing committee or program committee. Submissions will be evaluated based on overall quality and alignment with the workshop theme.

Notes:

Authors may choose to submit their work as anonymized (double-blind) or not anonymized (single-blind).
Submissions for both tracks will be accepted through PCS.
Please use the VGTC Conference Style template for submissions.
At least one author of accepted papers must register and attend in person.