CANDICE: Agentic Causal Disentanglement with Class Conditional Knowledge Integration for Long Tailed Domain Generalization

Midhat Urooj, Ayan Banerjee, Sandeep Gupta


Abstract
Deep learning models deployed in clinical settings face two major challenges: domain generalization (DG) and long-tailed (LT) recognition. DG requires learning domain-invariant features to ensure robustness across heterogeneous acquisition protocols and patient populations. However, we identify a fundamental trade-off: objectives that enforce domain invariance often suppress class-discriminative signals essential for long-tailed recognition.To address this, we propose the Agentic Causal Disentanglement (CANDICE) Framework, a modular architecture that integrates explicit clinical expertise from sonographers, radiologists, and specialists as a form of causal intervention. The framework combines clinical reasoning, causal representation learning, and automated pipeline construction to disentangle domain-invariant and class-discriminative features. By incorporating domain-specific causal knowledge, it effectively decouples the objectives of DG and LT learning. We evaluate CANDICE on 10 diverse medical imaging datasets spanning four modalities. The framework achieves an average performance improvement of 10.3% across both multi-domain and in-domain long-tailed tasks, demonstrating its effectiveness in handling distribution shifts while preserving minority class performance.
Anthology ID:
2026.findings-acl.2018
Volume:
Findings of the Association for Computational Linguistics: ACL 2026
Month:
July
Year:
2026
Address:
San Diego, California, United States
Editors:
Maria Liakata, Viviane P. Moreira, Jiajun Zhang, David Jurgens
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
40596–40624
Language:
URL:
https://aclanthology.org/2026.findings-acl.2018/
DOI:
10.18653/v1/2026.findings-acl.2018
Bibkey:
Cite (ACL):
Midhat Urooj, Ayan Banerjee, and Sandeep Gupta. 2026. CANDICE: Agentic Causal Disentanglement with Class Conditional Knowledge Integration for Long Tailed Domain Generalization. In Findings of the Association for Computational Linguistics: ACL 2026, pages 40596–40624, San Diego, California, United States. Association for Computational Linguistics.
Cite (Informal):
CANDICE: Agentic Causal Disentanglement with Class Conditional Knowledge Integration for Long Tailed Domain Generalization (Urooj et al., Findings 2026)
Copy Citation:
PDF:
https://aclanthology.org/2026.findings-acl.2018.pdf
Checklist:
 2026.findings-acl.2018.checklist.pdf