Trustworthy Foundation Models for Web Intelligence: Causal Perspectives and Challenges

Keynote Speakers (A-Z order)

Call for Papers

The rise of Foundation Models (FMs) presents unprecedented challenges for building reliable and trustworthy AI systems. This half-day workshop at WWW 2026 is dedicated to establishing Causal Reasoning as the essential, principled framework for ensuring the Trustworthiness of large-scale models.

We invite original research contributions that address the fundamental challenges of ensuring causality, fairness, transparency, robustness, and interpretability in FMs. We specifically encourage work that leverages causal inference for the analysis, diagnosis, and mitigation of trustworthiness issues, moving beyond correlation-based methods to establish true causal understanding.

Topics of Interest

We seek submissions that contribute to theoretical advancements, practical implementations, and ethical considerations in building causally grounded, trustworthy foundation models for the web. Topics include, but are not limited to:

• Causal Foundations for Trustworthy FMs
  • Causal Representation Learning and Disentanglement in Web Data.
  • Causal Discovery and Inference on large-scale, heterogeneous web graphs.
  • Integrating causal models to diagnose and explain FM behavior (C-XAI).
  • Understanding and mitigating confounding factors in web-scale training data.
• Trustworthiness in Web Applications
  • Fairness and Bias Mitigation (e.g., demographic bias, filter bubbles) in Recommender Systems, Search Engines, and Content Generation.
  • Robustness to Distributional Shifts (e.g., concept drift, adversarial attacks) in dynamic online environments.
  • Counterfactual Reasoning for personalized user modeling and decision-making on the Web.
  • Evaluation frameworks and metrics for causally grounded Trustworthy AI.
• Ethical and Societal Implications
  • Data and Model Governance for web-scale foundation models.
  • Responsible deployment and monitoring of FMs in high-stakes web domains.
  • Case studies and methodologies for identifying and explaining ethical risks.

Submission Guidelines

Submissions should be formatted according to the ACM style and submitted electronically through the workshop's submission portal. Neither the paper checklist nor a broader impact statement are required for workshop submissions.
We accept papers (no less than 4 pages) describing new research, preliminary results, or challenging open problems. Accepted papers will be presented as orals or posters.
All submissions must be original and should not have been published or be under review elsewhere.

Important Dates

Paper Submission Deadline	January 6, 2026
Notification of Acceptance	January 13, 2026
Camera-Ready Deadline	February 2, 2026
Workshop Date	April 13 to April 14 2026

All deadlines are 11:59 PM Anywhere on Earth (AoE)

Awards

Outstanding submissions will be recognized with Outstanding Paper Award and Outstanding Student Paper Award. Recipients of these prestigious honors will be invited to present their work in dedicated, oral, in-person sessions at the workshop.
Please note: Award eligibility requires in-person presentation at the workshop; failure to do so will result in the cancellation of the award.

Rising Star Award Announcement

The TrustFM Rising Star Award recognizes early-career researchers whose work advances trustworthy foundation models grounded. The award will be hosted by CausalTFM Workshop at WWW 2026, and two researchers will be selected. Awardees will receive certificates and be invited to deliver in-person spotlight talks at the workshop.

Scope and Vision

Modern web applications increasingly rely on large foundation models trained on complex, dynamic web data. While these models deliver impressive performance, they often operate as opaque correlation engines. This award focuses on early-career researchers who use causal perspectives to make such systems more reliable, robust, transparent, interpretables, fair, or accountable. We especially value work that connects theory and practice—for example, causal methods that drive improvements in web-scale systems, evaluation pipelines, or policy decisions. We warmly encourage applications from researchers from minority or underrepresented groups in artificial intelligence, machine learning, web, and data science communities.

Research Areas of Interest

We welcome applications working on (but not limited to) the following themes:

• Robustness of foundation models under distribution shifts in web environments
• Causal representation learning for foundation models on web-scale data
• Interpretability and explanation of model predictions
• Causality-driven debiasing and fairness for search, recommendation, ranking, and social platforms
• Causal reasoning with agents or tools built on top of foundation models for web intelligence
• Safety and alignment questions to diagnose or mitigate risks in web ecosystems
• Open-source libraries, datasets, or systems that use trustworthy methods for web-scale applications
• Causality-aware evaluation metrics, benchmarks, or tools for trustworthy foundation models

Eligibility and Requirements

• Senior PhD students enrolled in a PhD program before April 2023, or
• Researchers holding postdoctoral positions who obtained their PhD after December 2023

Application Materials

Applicants are required to submit the following via this form (except for recommendation letters):

1. CV, including a list of publications
2. Research statement (up to 2 pages, single column, excluding references) describing your research accomplishments and future directions
3. Two recommendation letters, to be uploaded by your referees before March 7, 2026 (AoE) via this form

Schedule (PST)

Organizers