[fix](README.md)

CuriousVLA_data/QA_sft_navsim_train_cot_fte142k_baseline_norm.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd16d83591b76f0336c0ea0406527e77fd53e2ffc4047d315b649c92317139f8
+size 766543664

README.md

@@ -1,22 +1,20 @@
-# CuriousVLA_dev
+# Curious-VLA-dev
 
-> Uploading...
+  This repository hosts the development release assets for Curious-VLA, including:
 
-## 1. Data
+  - SFT training data
+  - SFT-stage model weights
 
-### SFT
+  For full documentation, environment setup, evaluation, SFT training, and GRPO training, please refer to the main project:
 
-Recommendation:
+  - GitHub: https://github.com/Mashiroln/Curious-VLA
 
-1. baseline data + qwen2.5vl ✅
-2. golden data + qwen3vl ✅
+  ## Recommended Usage
 
-golden data + qwen2.5vl ❌
+  - baseline data + Qwen2.5-VL
+  - golden data + Qwen3-VL
 
+  ## Notes
 
-## 2. Model
-
-### SFT
-
-1. stage 2 model, which can train with GRPO directly.
-
+  - The SFT-stage model can be used as the starting point for GRPO training in the main Curious-VLA project.
+  - For RL training data, released token filters, and end-to-end scripts, see the main repository.