SAGA_Qwen3-8B

SAGA-finetuned Qwen3-VL-8B model for short-drama video understanding, trained on the M-Drama dataset (SFT + RL).

This model is introduced in Beyond Sparse Rewards: A New Benchmark and Structure-Aware Graph Alignment for Micro-Drama Understanding.
Code: github.com/qyx1121/MDrama_SAGA

Usage

from transformers import AutoProcessor, AutoModelForImageTextToText
import torch

model = AutoModelForImageTextToText.from_pretrained(
    "yixin1121/SAGA_Qwen3-8B", torch_dtype=torch.bfloat16, device_map="auto"
)
processor = AutoProcessor.from_pretrained("yixin1121/SAGA_Qwen3-8B")

Training

• Base model: Qwen3-VL-8B-Instruct
• Training data: M-Drama — 32,361 QA annotations over 8,102 short-drama video clips
• Pipeline: SFT followed by RL (DAPO-style, with graph-structure and semantic rewards)

License

CC BY-NC 4.0. This model is intended for research purposes only.