GOTZSL: Optimal Transport-Guided Graph-Aware Feature Alignment for Compositional Zero-Shot Learning

ACL ARR 2025 July Submission1401 Authors

Abstract:

Compositional Zero-Shot Learning (CZSL) aims to recognize unseen attribute-object compositions by generalizing from seen ones. Existing prompt-based methods often suffer from textual feature shift, while graph-based approaches are limited by static structures and lack compositional adaptability.

We propose GOTZSL: Optimal Transport-Guided Graph-Aware Feature Alignment for Compositional Zero-Shot Learning, a unified framework that integrates triple prompt tuning, a graph-based adapter, and compositional visual adaptation. GOTZSL encodes state, object, and pair prompts through triple-level text templates, refines them via a compositional graph aligned with LLM-derived anchors, and disentangles LoRA-adapted visual features using a dual-branch MLP module.

To improve consistency and generalization, we introduce a pairwise optimal transport loss and partial label smoothing over semantically related classes. Evaluated on UT-Zappos, MIT-States, and CGQA under both closed- and open-world CZSL settings, GOTZSL achieves state-of-the-art performance, demonstrating robust compositional reasoning.

Paper Type: Long

Research Area: Multimodality and Language Grounding to Vision, Robotics and Beyond

Research Area Keywords: image text matching;multimodality;cross-modal application;

Contribution Types: Approaches low compute settings-efficiency

Languages Studied: English

Reassignment Request Area Chair: This is not a resubmission

Reassignment Request Reviewers: This is not a resubmission

A1 Limitations Section: This paper has a limitations section.

A2 Potential Risks: N/A

B Use Or Create Scientific Artifacts: Yes

B1 Cite Creators Of Artifacts: N/A

B2 Discuss The License For Artifacts: N/A

B3 Artifact Use Consistent With Intended Use: N/A

B4 Data Contains Personally Identifying Info Or Offensive Content: N/A

B5 Documentation Of Artifacts: N/A

B6 Statistics For Data: Yes

B6 Elaboration: Section 4

C Computational Experiments: Yes

C1 Model Size And Budget: N/A

C2 Experimental Setup And Hyperparameters: Yes

C2 Elaboration: section 4

C3 Descriptive Statistics: Yes

C3 Elaboration: section 4

C4 Parameters For Packages: N/A

D Human Subjects Including Annotators: No

D1 Instructions Given To Participants: N/A

D2 Recruitment And Payment: N/A

D3 Data Consent: N/A

D4 Ethics Review Board Approval: N/A

D5 Characteristics Of Annotators: N/A

E Ai Assistants In Research Or Writing: No

E1 Information About Use Of Ai Assistants: N/A

Author Submission Checklist: yes

Submission Number: 1401