Qiaoyuan/qwen3-1p7b-polar-p1-lora

Qwen3-1.7B POLAR P1 Policy Inducer LoRA

This repository contains a LoRA adapter trained for minimal POLAR-Bench P1 policy induction.

Base model: Qwen/Qwen3-1.7B

Task: Given a source document, privacy policy, and task instruction, output:

{
  "scoring_targets": {
    "allowed_values": [],
    "do_not_disclose_values": []
  }
}
This is an adapter-only repository. Load it with PEFT on top of Qwen/Qwen3-1.7B.