HEAL @ CHI 2025
Human-centered Evaluation and
Auditing of Language Models

Yokohama, Japan | April 26-May 1 (TBD)

Submission Deadline: February 17, 2025 (AOE)

Submission Site

Overview

HEAL is back for CHI 2025! This workshop aims to address the current ''evaluation crisis'' in LLM research and practice by bringing together HCI and AI researchers and practitioners to rethink LLM evaluation and auditing from a human-centered perspective. The recent advancements in Large Language Models (LLMs) have significantly impacted numerous and will impact more, real-world applications. However, these models also pose significant risks to individuals and society. To mitigate these issues and guide future model development, responsible evaluation and auditing of LLMs are essential.

The CHI 2025 Workshop on Human-centered Evaluation and Auditing of Language Models (HEAL@CHI'25) will explore topics around understanding stakeholders' needs and goals with evaluation and auditing LMs, establishing human-centered evaluation and auditing methods, developing tools and resources to support these methods, building community, and fostering collaboration

Special Theme - Mind the Context: For this year’s HEAL, we introduce the theme of ''mind the context'' to encourage attendees to engage with specific contexts in LLM evaluation and auditing. This theme involves various topics: the usage contexts of LLMs (e.g., evaluating the capabilities and limitations of LLM applications in mental wellness care, or translation in high-stakes scenarios), the context of the evaluation/auditing itself (e.g., who are using LLM evaluation tools, and how should we design these tools with this context in mind?), and more. We purposefully leave ''context'' open for interpretation by participants, so to encourage diversity in how participants conceptualize and operationalize this key concept in LLM evaluation and auditing.

Key Information

Submission deadline: February 17, 2025 (AoE)

Notification of acceptance: March 17, 2025

Workshop date: April 26-May 1 (TBD)

Workshop location: Yokohama, Japan (Hybrid)

Contact: heal.workshop@gmail.com

Call for Participation

We welcome participants who work on topics related to supporting human-centered evaluation and auditing of language models. Interested participants will be asked to contribute a short paper to the workshop. Topics of interest include, but not limited to:

  • Empirical understanding of stakeholders' needs and goals of LLM evaluation and auditing
  • Human-centered evaluation and auditing methods for LLMs
  • Tools, processes, and guidelines for LLM evaluation and auditing
  • Discussion of regulatory measures and public policies for LLM auditing
  • Ethics in LLM evaluation and auditing
Special Theme: Mind the Context. We invite authors to engage with specific contexts in LLM evaluation and auditing. This theme could involve various topics: the usage contexts of LLMs (e.g., evaluating the capabilities and limitations of LLM applications in mental wellness care, or translation in high-stakes scenarios), the context of the evaluation/auditing itself (e.g., who are using LLM evaluation tools, and how should we design these tools with this context in mind?), and more. The term ''context'' is left open for interpretation, so to encourage diversity in how this this key concept is conceptualized and operationalized by workshop participants. Papers under this theme will be given a dedicated lightning talk session, as well as a special spotlight during the workshop's poster session.

Submission Format: 2 - 6 pages ACM double-column, excluding references.

Submission Types: Position papers, full or in-progress empirical studies, literature reviews, system demos, method descriptions, or encore of published work. The submission will be non-archival.

Review Process: Double-blind. Papers will be selected based on the quality of the submission and diversity of perspectives to allow for a meaningful exchange of knowledge between a broad range of stakeholders.

Templates: [Word] [LaTex] [Overleaf]

Notes:

  • We encourage authors who submit also to help with the review process.
  • For an encore submission, you do not need to anonymize the submission. Encore submissions will go through a jury review process.
  • Please use \documentclass[sigconf,anonymous]{acmart} for submission.
  • Please be aware of OpenReview's moderation policy for newly created profiles: new profiles created without an institutional email will go through a moderation process that can take up to two weeks, while new profiles created with an institutional email will be activated automatically.

→ Submission Site

Organizers