Update README.md

README.md

@@ -80,26 +80,21 @@ JiuZhou outperforms GPT-3.5 in objective tasks:
     <img src="image/objective_score.png" width="800"/>
     <br>
 </p>
-
-JiuZhou also scores higher than ClimateChat across six criteria in subjective tasks:
+JiuZhou also scores higher than JiuZhou across six criteria in subjective tasks:
 <p align="center">
     <br>
     <img src="image/subjective_score.png" width="800"/>
     <br>
 </p>
-
 ### General Ability
-
-We evaluate the performance of Chinese-Mistral-7B using three benchmark datasets: C-Eval, CMMLU, and MMLU.<br>
+We evaluate the performance of JiuZhou using three benchmark datasets: C-Eval, CMMLU, and MMLU.<br>
 Compared to other variants of Llama and Mistral models, JiuZhou shows outstanding performance:
 <p align="center">
     <br>
     <img src="image/general_score.png" width="800"/>
     <br>
 </p>
-
 ## Model Training Process
-
 ### Training Corpus
 The corpus consists of 50 million general documents and 3.4 million geoscience-related documents.
 <p align="center">
@@ -107,7 +102,6 @@ The corpus consists of 50 million general documents and 3.4 million geoscience-r
     <img src="image/JiuZhou-Corpus.png" width="800"/>
     <br>
 </p>
-
 ### Training Framework
 We use the JiuZhou-Framework proposed in this study.
 <p align="center">
@@ -115,7 +109,6 @@ We use the JiuZhou-Framework proposed in this study.
     <img src="image/JiuZhou-Framework.png" width="800"/>
     <br>
 </p>
-
 ### Two-stage Pre-adaptation Pre-training (TSPT)
 TSPT improves the efficiency of using limited geoscience data and overcomes some of the technical bottlenecks in continual pretraining for LLMs.<br>
 The difference between TSPT and single-stage training algorithms:
@@ -130,8 +123,6 @@ Comparison of TSPT and one-stage pre-training algorithm performance:
     <img src="image/TSPT_score.png" width="800"/>
     <br>
 </p>
-
-
 ## Model Training Code
 We use [LLaMA-Factory](https://github.com/hiyouga/LLaMA-Factory) to fine-tune JiuZhou.