Update README.md
Browse files
README.md
CHANGED
|
@@ -34,7 +34,7 @@ base_model:
|
|
| 34 |
|
| 35 |
**AntAngelMed is Officially Open Source! 🚀**
|
| 36 |
|
| 37 |
-
**AntAngelMed**, jointly developed by the **Health Information Center of Zhejiang Province, Ant
|
| 38 |
|
| 39 |
# Core Highlights
|
| 40 |
|
|
@@ -76,7 +76,7 @@ base_model:
|
|
| 76 |
AntAngelMed employs a carefully designed three-stage training process to deeply integrate general capabilities with medical expertise:
|
| 77 |
|
| 78 |
+ **Continual Pre-Training:** Based on Ling-flash-2.0, AntAngelMed is continually pre-trained with large-scale, high-quality medical corpora (encyclopedias, web text, academic publications), injecting profound domain and world knowledge.
|
| 79 |
-
+ **Supervised Fine-Tuning (SFT):** A multi-source and heterogeneous high-quality instruction dataset is constructed at this stage. General data (math, programming, logic) strengthen core chain-of-thought capabilities of
|
| 80 |
+ **Reinforcement Learning (RL):** Using the [**GRPO**](https://arxiv.org/pdf/2402.03300) algorithm and task-specific reward models, RL precisely shapes model behavior—emphasizing empathy, structural clarity, and safety boundaries, and encouraging evidence-based reasoning on complex cases to reduce hallucinations and improve accuracy.
|
| 81 |
|
| 82 |
|
|
|
|
| 34 |
|
| 35 |
**AntAngelMed is Officially Open Source! 🚀**
|
| 36 |
|
| 37 |
+
**AntAngelMed**, jointly developed by the **Health Information Center of Zhejiang Province, Ant Healthcare, and Zhejiang Anzhen'er Medical Artificial Intelligence Technology Co., Ltd.(浙江省安诊儿医学人工智能科技有限公司),** is the largest and most powerful open-source medical language model to date.
|
| 38 |
|
| 39 |
# Core Highlights
|
| 40 |
|
|
|
|
| 76 |
AntAngelMed employs a carefully designed three-stage training process to deeply integrate general capabilities with medical expertise:
|
| 77 |
|
| 78 |
+ **Continual Pre-Training:** Based on Ling-flash-2.0, AntAngelMed is continually pre-trained with large-scale, high-quality medical corpora (encyclopedias, web text, academic publications), injecting profound domain and world knowledge.
|
| 79 |
+
+ **Supervised Fine-Tuning (SFT):** A multi-source and heterogeneous high-quality instruction dataset is constructed at this stage. General data (math, programming, logic) strengthen core chain-of-thought capabilities of AngAngelMed, while medical scenarios (doctor–patient Q&A, diagnostic reasoning, safety/ethics) provide deep adaptation for improved clinical performance.
|
| 80 |
+ **Reinforcement Learning (RL):** Using the [**GRPO**](https://arxiv.org/pdf/2402.03300) algorithm and task-specific reward models, RL precisely shapes model behavior—emphasizing empathy, structural clarity, and safety boundaries, and encouraging evidence-based reasoning on complex cases to reduce hallucinations and improve accuracy.
|
| 81 |
|
| 82 |
|