Modelpublic

You-En/NYCU-IAlI-DL2026-LLM1-SFT_on_Answer-Only_Data

sourceHugging Faceupdated 3mo agoView on Hugging Face

0likes8downloads

Model Card

NYCU-IAII-DL2026 LLM1 SFT on Answer-Only Data

This repository contains the final LoRA adapter for the NYCU-IAII-DL2026 LLM #1 task:

SFT on Answer-Only Data

The adapter is fine-tuned from:

text

Qwen/Qwen2.5-7B-Instruct

This repository does not contain a full merged model. It contains a PEFT / LoRA adapter that should be loaded on top of the base model.

Model Details

—Base model: Qwen/Qwen2.5-7B-Instruct
—Fine-tuning method: Supervised fine-tuning
—Training format: Answer-only multiple-choice QA
—Adapter method: LoRA / PEFT
—Quantization during training: 4-bit quantization
—Framework: PyTorch, Transformers, PEFT
—Expected output: one of A, B, C, or D

Limitations

This adapter is trained specifically for course multiple-choice answer-only QA. It may not generalize well to open-ended dialogue or general reasoning tasks.