Add pipeline tag and library metadata

README.md

@@ -1,7 +1,9 @@
 ---
-license: apache-2.0
 base_model:
 - Qwen/Qwen2.5-VL-3B-Instruct
+license: apache-2.0
+library_name: transformers
+pipeline_tag: image-text-to-text
 ---
 
 # SpatialEvo: Self-Evolving Spatial Intelligence via Deterministic Geometric Environments
@@ -21,12 +23,11 @@ base_model:
 
 ## SpatialEvo-3B
 
-
 This repository contains **SpatialEvo-3B**, introduced in [SpatialEvo: Self-Evolving Spatial Intelligence via Deterministic Geometric Environments](https://arxiv.org/abs/2604.14144).
 
 ## Model Description
 
-SpatialEvo-3B is fine-tuned from **Qwen2.5-VL-3B-Instruct** using the SpatialEvo self-evolving framework. Instead of relying on manually annotated datasets or model voting to construct pseudo-labels, SpatialEvo leverages a **Deterministic Geometric Environment (DGE)** that programmatically computes exact ground truth from 3D point clouds and camera poses, enabling zero-noise online reinforcement learning across 16 spatial reasoning task categories.
+SpatialEvo-3B is fine-tuned from **Qwen2.5-VL-3B-Instruct** using the SpatialEvo self-evolving framework. Instead of relying on manually annotated datasets or model consensus to construct pseudo-labels, SpatialEvo leverages a **Deterministic Geometric Environment (DGE)** that programmatically computes exact ground truth from 3D point clouds and camera poses, enabling zero-noise online reinforcement learning across 16 spatial reasoning task categories.
 
 A single shared-parameter policy co-evolves as both a **Questioner** and a **Solver** under GRPO optimization, while a lightweight **Task Scheduler** drives adaptive curriculum learning based on historical accuracy — without any manual stage design or human annotation.