laonML
/

LaonA2_VL_3B

Zero-Shot Object Detection

Model card Files Files and versions

LaonA2 VL 3B

LaonA2 VL 3B는 Qwen 2.5 VL 3B 기반의 향상된 비전-언어 모델입니다. VLM-R1 강화학습을 통해 REC(Referring Expression Comprehension) 성능이 개선되었습니다.

cite: arxiv.org/abs/2504.07615

Downloads last month: 16

Inference Providers NEW

Zero-Shot Object Detection

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for laonML/LaonA2_VL_3B

Base model

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

Finetuned

(892)

this model

Dataset used to train laonML/LaonA2_VL_3B

Paper for laonML/LaonA2_VL_3B

VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model

Paper • 2504.07615 • Published Apr 10, 2025 • 36