You can deploy this model and start earning money today!

JosephZ/
R1-SGG-7B-PSG

Name: R1-SGG-7B-PSG
Availability: InStock

scene-graph-generation

image-text-to-text

JosephZ/vg150_train_sgg_prompt

2504.13617

Qwen/Qwen2-VL-7B-Instruct

finetune

apache-2.0

text-generation-inference

endpoints_compatible

Views

No views yet

Model Description

An end-to-end multimodal LLM for Scene Graph Generation (SGG), which was introduced in Compile Scene Graphs with Reinforcement Learning

R1-SGG: Compile Scene Graphs with Reinforcement Learning

Structured Visual Reasoning with Multimodal LLMs and Reinforcement Learning

Paper

License

Hugging Face

🚀 Update

✅
Hugging Face
R1-SGG-7B, R1-SGG-Zero-7B
✅ Support PSG dataset (bbox format only, not Panoptic)
✅ Updated loss implementation
✅ Always use custom_per_device_train_batch_size instead of per_device_train_batch_size for faster sampling under gradient accumulation
⚠️ Current loss implementation might still be affected by gradient accumulation: trl issue #3021

🛠️ Setup Environment

bash install.sh

Main dependencies:

bash

- torch == 2.5.0 or 2.5.1 (cu124, optional)
- transformers (supports Qwen2VL, Qwen2.5VL)
- trl
- vLLM

📚 Dataset

Load preprocessed datasets via:

python

from datasets import load_dataset

db_train = load_dataset("JosephZ/vg150_train_sgg_prompt")["train"]
db_val = load_dataset("JosephZ/vg150_val_sgg_prompt")["train"]

or for PSG:

python

db_train = load_dataset("JosephZ/psg_train_sg")["train"]  # keys: image_id, image, objects, relationships
db_val = load_dataset("JosephZ/psg_test_sg")["train"]

We transformed VG150 into HuggingFace Datasets format with keys:

image_id
image
prompt_open
prompt_close
objects
relationships

🔥 Supported Models

Qwen/Qwen2-VL-2B-Instruct
Qwen/Qwen2-VL-7B-Instruct
Qwen/Qwen2.5-VL-3B-Instruct
Qwen/Qwen2.5-VL-7B-Instruct

🏋️‍♂️ Training

Training with Supervised Fine-Tuning (SFT)

For SLURM users:

sbatch scripts/sft/7B_sgg.sh

For local machines:

bash scripts/sft_local/7B_sgg.sh

⏱️ Approximate training time:

2B models: ~4 hours (4×A100 SXM4 GPUs)
7B models: ~10 hours (4×A100 SXM4 GPUs)

Training with Reinforcement Learning (GRPO)

** Update (11/05/2025): to use "Hard Recall"**:

--reward_funcs format_reward edge_hard_reward

For A100 GPUs:

sbatch scripts/grpo/train_a100_2B.sh

(12 hours on 16×A100 GPUs)

For GH200 GPUs:

sbatch scripts/grpo/train_gh200.sh

(16 hours on 16×GH200 GPUs)

For clusters with many RTX_3090/4090 GPUs:

sbatch scripts/grpo/train_fused.sh

Training 7B models on 24GB cards is possible with Zero3, but slow due to communication bottlenecks.
(Fun fact: training with 120×RTX_4090 is crazy but severely limited by communication latency.)

💡 Recommended learning rate: 6e-7.

🧪 Inference and Evaluation

Inference with SFT-trained models:

bash scripts/inference/run_sgg_inference.sh $DATASET $MODEL_NAME $OUTPUT_DIR

For models trained with predefined categories, add true:

bash scripts/inference/run_sgg_inference.sh $DATASET $MODEL_NAME $OUTPUT_DIR true

Inference with GRPO-trained models:

bash scripts/inference/run_sgg_inference.sh $DATASET $MODEL_NAME $OUTPUT_DIR false/true true

Evaluation: