Update README.md

README.md

@@ -6,4 +6,21 @@ pipeline_tag: image-to-text
 tags:
 - multimodal
 - image caption
----
+library_name: transformers
+---
+
+# CapRL-3B
+## Introduction
+We are excited to introduce CapRL-3B, a lightweight 3B captioner that achieves perception capabilities comparable to Qwen2.5-VL-72B.
+
+This is the first study of applying Reinforcement Learning with Verifiable Rewards for the
+open-ended and subjective image captioning task. Unlike traditional Supervised Fine-Tuning, which
+can lead to models memorizing a limited set of annotated captions, our method allows the model to
+explore and generate a broader range of creative and general descriptions.
+
+CapRL is a new training paradigm featuring a decoupled two-stage pipeline. The initial
+stage uses LVLMs to generate rich and accurate captions. Subsequently, the second stage evaluates
+caption quality by using a vision-only LLM to perform the QA task. We also created a specific QA
+curation pipeline to ensure the quality of the questions and answers used for the second stage.
+
+By employing our CapRL training framework, initializing with the Qwen2.5-VL-3B model, and using a carefully filtered 75K QA dataset as the training set, we obtained a highly capable captioner, CapRL-3B.