VUL337-OL at LLMs4OL 2026 (Task A & Task B)

[caravel23]

Hello,

We are submitting the script-based system integration packages for team VUL337-OL.

Task A - Flagship

Our submitted Full configuration implements progressive evidence-constrained ontology induction from raw text. It consists of evidence-grounded endpoint discovery, evidence-guided directed pair proposal, subject-wise competitive parent selection, and retrieval-calibrated graph revision.

Task B - Reuse

Our system performs ontology-local zero-shot parent and relation selection using the supplied text and partial ontology. It does not require model training or embedding retrieval.

Both packages include:

• a primary main.py entry point;
• installation and execution instructions in README.md;
• pinned dependencies in requirements.txt;
• offline tests in unittest.py;
• an MIT license.

The two integration packages are attached separately for Task A and Task B.

Best regards,
VUL337-OL

VUL337-OL_TaskA_Flagship.zip

VUL337-OL_TaskB_Reuse.zip

[HamedBabaei]

Dear @caravel23, thanks for the submission. I will mark your submission as completed. We will use this submission to do the integration.

[HamedBabaei]

System Integration Submission Checklist

Team Name: VUL337-OL
Challenge Variant: LLMs4OL-2026 Flagship and Reuse
Submission Type: Script

---

General Requirements

• Team name and challenge variant are included in the issue (or PR) title.
• The submission is complete and executable.
• Unit tests are included for the core system components.
• A methodological diagram is included.
• A link to the project GitHub repository is included.

---

Script-Based Submission

Project Structure

• All source files are placed in the project root (no nested directories).
• main.py is provided as the primary entry point.
• Additional entry files (main1.py, main2.py, etc.) are provided if multiple variants exist.
• requirements.txt is included.
• README.md is included.
• unittest.py is included.

Documentation

• README explains how to run the system.
• README briefly describes the purpose of each script.
• README documents multiple entry points (if applicable).

Unit Testing

• unittest.py covers the critical modules.

Submission

• Project is packaged as a ZIP archive.

---

Additional Information

• Methodological Diagram:
• Project Repository:
• Additional Notes:

This comment will be updated based on integration progress and checking the submissions. This will be on the OntoLearner Team, and there will be no need for the authors to do so. Unless you (as a co-author) want to add more information or want to be involved in the integration process.