diff --git a/README.md b/README.md
index 58a4fe20d9df7d7a6c6b9e9343232e37baac6c20..61bc70bbb67d02129744b852f9d6ef5622f750c0 100644
--- a/README.md
+++ b/README.md
@@ -4,62 +4,6 @@ DeepSparkHub甄选上百个应用算法和模型，覆盖AI和通用计算各领
 
 ## 模型列表
 
-- Computer Vision
-
-  - [Classification](#classification)
-  - [Face Detection](#face-detection)
-  - [Face Recognition](#face-recognition)
-  - [Instance Segmentation](#instance-segmentation)
-  - [Knowledge Distillation](#knowledge-distillation)
-  - [Network Pruning](#network-pruning)
-  - [Object Detection](#object-detection)
-  - [3D Object Detection](#3d-object-detection)
-  - [OCR](#ocr)
-  - [Point Cloud](#point-cloud)
-  - [Pose Estimation](#pose-estimation)
-  - [Self-Supervised Learning](#self-supervised-learning)
-  - [Semantic Segmentation](#semantic-segmentation)
-  - [Super Resolution](#super-resolution)
-  - [Tracking](#tracking)
-  - [Traffic Forecast](#traffic-forecast)
-
-- Graph Neural Network (GNN)
-
-  - [Graph Attention](#graph-attention)
-  - [Node Classification](#node-classification)
-  - [Text Classification](#text-classification)
-
-- High Performance Computing (HPC)
-
-  - [Molecular Dynamics](#molecular-dynamics)
-
-- [Multimodal](#multimodal)
-
-- Natural Language Processing (NLP)
-
-  - [Cloze Test](#cloze-test)
-  - [Dialogue Generation](#dialogue-generation)
-  - [Language Modeling](#language-modeling)
-  - [Large Language Model (LLM)](#large-language-model-llm)
-  - [Text Correction](#text-correction)
-  - [Translation](#translation)
-
-- Recommendation
-
-  - [Collaborative Filtering](#collaborative-filtering)
-  - [Click Through Rate](#click-through-rate)
-
-- [Reinforcement Learning](#reinforcement-learning)
-
-- Speech
-
-  - [Speech Recognition](#speech-recognition)
-  - [Speech Synthesis](#speech-synthesis)
-
-- [3D Reconstruction](#3d-reconstruction)
-
---------
-
 ### Computer Vision
 
 #### Classification
@@ -197,6 +141,7 @@ DeepSparkHub甄选上百个应用算法和模型，覆盖AI和通用计算各领
 [Faster R-CNN](cv/detection/fasterrcnn/pytorch/README.md)  | PyTorch | COCO
 [FCOS](cv/detection/fcos/paddlepaddle/README.md)  | PaddlePaddle | COCO
 [FCOS](cv/detection/fcos/pytorch/README.md)  | PyTorch | COCO
+[Mamba-YOLO](cv/detection/mamba_yolo/pytorch/README.md)  | PyTorch | COCO
 [Mask R-CNN](cv/detection/maskrcnn/pytorch/README.md)  | PyTorch | COCO
 [Mask R-CNN](cv/detection/maskrcnn/paddlepaddle/README.md)  | PaddlePaddle | COCO
 [OC_SORT](cv/detection/oc_sort/paddlepaddle/README.md)  | PaddlePaddle | MOT17
@@ -388,6 +333,14 @@ DeepSparkHub甄选上百个应用算法和模型，覆盖AI和通用计算各领
 -------- | ------ | ----
 [Water/se_e2_a](hpc/molecular_dynamics/water_se_e2_a/tensorflow/README.md)  | TensorFlow (DeePMD-kit) | data_water
 
+###  Methodology
+
+#### Kolmogorov-Arnold Networks
+
+模型名称 | 框架 | 数据集
+-------- | ------ | ----
+[KAN](methodology/kolmogorov_arnold_networks/kan/pytorch/README.md)  | PyTorch | -
+
 ### Multimodal
 
 模型名称 | 框架 | 数据集
@@ -396,7 +349,7 @@ DeepSparkHub甄选上百个应用算法和模型，覆盖AI和通用计算各领
 [CLIP](multimodal/Language-Image_Pre-Training/clip/pytorch/README.md)  | PyTorch | CIFAR100
 [ControlNet](multimodal/diffusion/ControlNet/README.md)  | PyTorch | Fill50K
 [DDPM](multimodal/diffusion/ddpm/README.md)  | PyTorch | CIFAR-10
-[LLaVA](multimodal/llava/pytorch/README.md)  | PyTorch | LLaVA-Pretrain
+[LLaVA 1.5](multimodal/llava/pytorch/README.md)  | PyTorch | LLaVA-Pretrain
 [L-Verse](multimodal/Language-Image_Pre-Training/L-Verse/pytorch/README.md)  | PyTorch | ImageNet
 [Stable Diffusion 1.4](multimodal/diffusion/stable-diffusion/training/README.md)  | PyTorch | pokemon-images
 [Stable Diffusion 1.5](multimodal/diffusion/stable-diffusion/sd_1.5/README.md)  | PyTorch | pokemon-images
@@ -445,6 +398,7 @@ DeepSparkHub甄选上百个应用算法和模型，覆盖AI和通用计算各领
 [ChatGLM-6B](nlp/llm/chatglm-6b/deepspeed/README.md)  | PyTorch (DeepSpeed) | ADGEN & chatglm-6b
 [ChatGLM2-6B SFT](nlp/llm/ChatGLM2-6b-sft/README.md)  | PyTorch (DeepSpeed) | ADGEN & chatglm2-6b
 [ChatGLM3-6B](nlp/llm/chatglm3-6b/deepspeed/finetune_demo/README.md)  | PyTorch (DeepSpeed) | ADGEN & chatglm3-6b
+[DeepSeekMoE 7B](nlp/llm/deepseek_moe_7b/colossalai/README.md)  | PyTorch (ColossalAI) | deepseek-moe-16b-base
 [Llama-7B](nlp/llm/llama-7b/colossalai/README.md)  | PyTorch (Colossal-AI) | llama-7b-hf
 [Llama2-7B](nlp/llm/llama2-7b/megatron-deepspeed/README.md)  | PyTorch (Megatron-DeepSpeed) | Bookcorpus
 [Llama2-7B Reward Model Finetuning](nlp/llm/llama2-7b_reward_sft/deepspeed/README.md)  | PyTorch (DeepSpeed) | Dahoas/rm-static
@@ -452,11 +406,12 @@ DeepSparkHub甄选上百个应用算法和模型，覆盖AI和通用计算各领
 [Llama2-7B SFT](nlp/llm/llama2-7b_sft/megatron-deepspeed/README.md)  | PyTorch (Megatron-DeepSpeed) | gpt_small-117M
 [Llama2-13B](nlp/llm/llama2-13b/megatron-deepspeed/README.md)  | PyTorch (Megatron-DeepSpeed) | Bookcorpus
 [Llama2-34B](nlp/llm/llama2-34b/megatron-deepspeed/README.md)  | PyTorch (Megatron-DeepSpeed) | Bookcorpus
-[Llama3-8B](nlp/llm/llama3-8b/megatron-deepspeed/README.md)  | PyTorch (Megatron-DeepSpeed) | Bookcorpus
+[Llama3-8B](nlp/llm/llama3_8b/megatron-deepspeed/README.md)  | PyTorch (Megatron-DeepSpeed) | Bookcorpus
+[Llama3-8B SFT](nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/README.md)  | PyTorch (ColossalAI) | school_math_0.25M
 [QWen-7B](nlp/llm/qwen-7b/firefly/README.md)  | PyTorch (Firefly) | qwen-7b
 [QWen1.5-7B](nlp/llm/qwen1.5-7b/firefly/README.md)  | PyTorch (Firefly) | school_math
 [QWen1.5-14B](nlp/llm/qwen1.5-14b/firefly/README.md)  | PyTorch (Firefly) | school_math
-[QWen2.5-7B](nlp/llm/qwen2.5-7b/LLaMA-Factory/README.md)  | PyTorch (LLaMA-Factory) | qwen2.5-7b
+[Qwen2.5-7B SFT](nlp/llm/qwen2.5-7b/LLaMA-Factory/README.md)  | PyTorch (LLaMA-Factory) | qwen2.5-7b
 
 #### Text Correction
 
diff --git a/dl/kan/README.md b/methodology/kolmogorov_arnold_networks/kan/pytorch/README.md
similarity index 91%
rename from dl/kan/README.md
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/README.md
index fd10c9351f9e4fda8981d71eee8213dbd2944979..ec6491e9e7a8102d464fef01aa5d01fbf3bd319e 100644
--- a/dl/kan/README.md
+++ b/methodology/kolmogorov_arnold_networks/kan/pytorch/README.md
@@ -1,23 +1,22 @@
 # KAN
 
 ## Model description
-Kolmogorov-Arnold Networks (KANs) are promising alternatives of Multi-Layer Perceptrons (MLPs). KANs have strong mathematical foundations just like MLPs: MLPs are based on the universal approximation theorem, while KANs are based on Kolmogorov-Arnold representation theorem. KANs and MLPs are dual: KANs have activation functions on edges, while MLPs have activation functions on nodes. This simple change makes KANs better (sometimes much better!) than MLPs in terms of both model accuracy and interpretability. 
 
+Kolmogorov-Arnold Networks (KANs) are promising alternatives of Multi-Layer Perceptrons (MLPs). KANs have strong mathematical foundations just like MLPs: MLPs are based on the universal approximation theorem, while KANs are based on Kolmogorov-Arnold representation theorem. KANs and MLPs are dual: KANs have activation functions on edges, while MLPs have activation functions on nodes. This simple change makes KANs better (sometimes much better!) than MLPs in terms of both model accuracy and interpretability.
 
 ## Run
-```shell
-$ pip3 install -r requirements.txt
-$ bash ./run_train.sh
 
+```sh
+pip3 install -r requirements.txt
+bash ./run_train.sh
 ```
 
 ## Result
+
 | Model       | Training speed   |
 |-------------|------------------|
 | KAN         | 6490 samples/sec |
 
-
 ## Reference
 
 - [pykan](https://github.com/KindXiaoming/pykan/tree/master)
-
diff --git a/dl/kan/__init__.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/__init__.py
similarity index 100%
rename from dl/kan/__init__.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/__init__.py
diff --git a/dl/kan/kan/.ipynb_checkpoints/KANLayer-checkpoint.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/KANLayer-checkpoint.py
similarity index 100%
rename from dl/kan/kan/.ipynb_checkpoints/KANLayer-checkpoint.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/KANLayer-checkpoint.py
diff --git a/dl/kan/kan/.ipynb_checkpoints/LBFGS-checkpoint.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/LBFGS-checkpoint.py
similarity index 100%
rename from dl/kan/kan/.ipynb_checkpoints/LBFGS-checkpoint.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/LBFGS-checkpoint.py
diff --git a/dl/kan/kan/.ipynb_checkpoints/MLP-checkpoint.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/MLP-checkpoint.py
similarity index 100%
rename from dl/kan/kan/.ipynb_checkpoints/MLP-checkpoint.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/MLP-checkpoint.py
diff --git a/dl/kan/kan/.ipynb_checkpoints/MultKAN-checkpoint.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/MultKAN-checkpoint.py
similarity index 100%
rename from dl/kan/kan/.ipynb_checkpoints/MultKAN-checkpoint.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/MultKAN-checkpoint.py
diff --git a/dl/kan/kan/.ipynb_checkpoints/Symbolic_KANLayer-checkpoint.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/Symbolic_KANLayer-checkpoint.py
similarity index 100%
rename from dl/kan/kan/.ipynb_checkpoints/Symbolic_KANLayer-checkpoint.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/Symbolic_KANLayer-checkpoint.py
diff --git a/dl/kan/kan/.ipynb_checkpoints/__init__-checkpoint.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/__init__-checkpoint.py
similarity index 100%
rename from dl/kan/kan/.ipynb_checkpoints/__init__-checkpoint.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/__init__-checkpoint.py
diff --git a/dl/kan/kan/.ipynb_checkpoints/compiler-checkpoint.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/compiler-checkpoint.py
similarity index 100%
rename from dl/kan/kan/.ipynb_checkpoints/compiler-checkpoint.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/compiler-checkpoint.py
diff --git a/dl/kan/kan/.ipynb_checkpoints/experiment-checkpoint.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/experiment-checkpoint.py
similarity index 100%
rename from dl/kan/kan/.ipynb_checkpoints/experiment-checkpoint.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/experiment-checkpoint.py
diff --git a/dl/kan/kan/.ipynb_checkpoints/feynman-checkpoint.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/feynman-checkpoint.py
similarity index 100%
rename from dl/kan/kan/.ipynb_checkpoints/feynman-checkpoint.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/feynman-checkpoint.py
diff --git a/dl/kan/kan/.ipynb_checkpoints/hypothesis-checkpoint.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/hypothesis-checkpoint.py
similarity index 100%
rename from dl/kan/kan/.ipynb_checkpoints/hypothesis-checkpoint.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/hypothesis-checkpoint.py
diff --git a/dl/kan/kan/.ipynb_checkpoints/spline-checkpoint.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/spline-checkpoint.py
similarity index 100%
rename from dl/kan/kan/.ipynb_checkpoints/spline-checkpoint.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/spline-checkpoint.py
diff --git a/dl/kan/kan/.ipynb_checkpoints/utils-checkpoint.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/utils-checkpoint.py
similarity index 100%
rename from dl/kan/kan/.ipynb_checkpoints/utils-checkpoint.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/.ipynb_checkpoints/utils-checkpoint.py
diff --git a/dl/kan/kan/KANLayer.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/KANLayer.py
similarity index 100%
rename from dl/kan/kan/KANLayer.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/KANLayer.py
diff --git a/dl/kan/kan/LBFGS.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/LBFGS.py
similarity index 100%
rename from dl/kan/kan/LBFGS.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/LBFGS.py
diff --git a/dl/kan/kan/MLP.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/MLP.py
similarity index 100%
rename from dl/kan/kan/MLP.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/MLP.py
diff --git a/dl/kan/kan/MultKAN.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/MultKAN.py
similarity index 100%
rename from dl/kan/kan/MultKAN.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/MultKAN.py
diff --git a/dl/kan/kan/Symbolic_KANLayer.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/Symbolic_KANLayer.py
similarity index 100%
rename from dl/kan/kan/Symbolic_KANLayer.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/Symbolic_KANLayer.py
diff --git a/dl/kan/kan/__init__.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/__init__.py
similarity index 100%
rename from dl/kan/kan/__init__.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/__init__.py
diff --git a/dl/kan/kan/assets/img/mult_symbol.png b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/assets/img/mult_symbol.png
similarity index 100%
rename from dl/kan/kan/assets/img/mult_symbol.png
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/assets/img/mult_symbol.png
diff --git a/dl/kan/kan/assets/img/sum_symbol.png b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/assets/img/sum_symbol.png
similarity index 100%
rename from dl/kan/kan/assets/img/sum_symbol.png
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/assets/img/sum_symbol.png
diff --git a/dl/kan/kan/compiler.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/compiler.py
similarity index 100%
rename from dl/kan/kan/compiler.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/compiler.py
diff --git a/dl/kan/kan/experiment.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/experiment.py
similarity index 100%
rename from dl/kan/kan/experiment.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/experiment.py
diff --git a/dl/kan/kan/experiments/experiment1.ipynb b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/experiments/experiment1.ipynb
similarity index 100%
rename from dl/kan/kan/experiments/experiment1.ipynb
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/experiments/experiment1.ipynb
diff --git a/dl/kan/kan/feynman.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/feynman.py
similarity index 100%
rename from dl/kan/kan/feynman.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/feynman.py
diff --git a/dl/kan/kan/hypothesis.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/hypothesis.py
similarity index 100%
rename from dl/kan/kan/hypothesis.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/hypothesis.py
diff --git a/dl/kan/kan/spline.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/spline.py
similarity index 100%
rename from dl/kan/kan/spline.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/spline.py
diff --git a/dl/kan/kan/utils.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/kan/utils.py
similarity index 100%
rename from dl/kan/kan/utils.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/kan/utils.py
diff --git a/dl/kan/requirements.txt b/methodology/kolmogorov_arnold_networks/kan/pytorch/requirements.txt
similarity index 100%
rename from dl/kan/requirements.txt
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/requirements.txt
diff --git a/dl/kan/run_train.sh b/methodology/kolmogorov_arnold_networks/kan/pytorch/run_train.sh
similarity index 100%
rename from dl/kan/run_train.sh
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/run_train.sh
diff --git a/dl/kan/train_kan.py b/methodology/kolmogorov_arnold_networks/kan/pytorch/train_kan.py
similarity index 100%
rename from dl/kan/train_kan.py
rename to methodology/kolmogorov_arnold_networks/kan/pytorch/train_kan.py
diff --git a/nlp/llm/deepseek_moe_7b/ColossalAI/README.md b/nlp/llm/deepseek_moe_7b/colossalai/README.md
similarity index 80%
rename from nlp/llm/deepseek_moe_7b/ColossalAI/README.md
rename to nlp/llm/deepseek_moe_7b/colossalai/README.md
index 1cbc5ba4754b006136161f9b05e53d954d53532a..4a02e88a184721793ac96959fd24402e4da489cf 100644
--- a/nlp/llm/deepseek_moe_7b/ColossalAI/README.md
+++ b/nlp/llm/deepseek_moe_7b/colossalai/README.md
@@ -1,8 +1,10 @@
-# Colossal-AI LLaMA-7B
+# DeepSeekMoE 7B (ColossalAI)
 
 ## Model description
+
+DeepSeekMoE 7B is a variant of the 16B model.
+
 DeepSeekMoE 16B is a Mixture-of-Experts (MoE) language model with 16.4B parameters. It employs an innovative MoE architecture, which involves two principal strategies: fine-grained expert segmentation and shared experts isolation.
-DeepSeekMoE 7B is a variant of the 16B model. 
 
 ## Step 1: Install
 
@@ -11,14 +13,16 @@ Firstly, you should ensure that ColossalAI is installed in the environment. Gene
 ## Step 2: Prepare model and config
 
 Get "deepseek-moe-16b-base" models and config file from huggingface or other place, and mv it to "/home/model_zoos/nlp/deepseek-moe-16b-base".
-One recommended link: "https://huggingface.co/deepseek-ai/deepseek-moe-16b-base/tree/main".
+One recommended link: "<https://huggingface.co/deepseek-ai/deepseek-moe-16b-base/tree/main>".
 
 ## Step 3: Training
+
 ```bash
-$ bash deepseek_7b_pretrain.sh
+bash deepseek_moe_7b_pretrain.sh
 ```
 
 ## Results
+
 | Model              | Training speed     |
 |--------------------|--------------------|
 | deepseek-moe-7b    |  6.85 samples/sec  |
diff --git a/nlp/llm/deepseek_moe_7b/ColossalAI/benchmark.py b/nlp/llm/deepseek_moe_7b/colossalai/benchmark.py
similarity index 100%
rename from nlp/llm/deepseek_moe_7b/ColossalAI/benchmark.py
rename to nlp/llm/deepseek_moe_7b/colossalai/benchmark.py
diff --git a/nlp/llm/deepseek_moe_7b/ColossalAI/data_utils.py b/nlp/llm/deepseek_moe_7b/colossalai/data_utils.py
similarity index 100%
rename from nlp/llm/deepseek_moe_7b/ColossalAI/data_utils.py
rename to nlp/llm/deepseek_moe_7b/colossalai/data_utils.py
diff --git a/nlp/llm/deepseek_moe_7b/ColossalAI/deepseek_moe_7b_pretrain.sh b/nlp/llm/deepseek_moe_7b/colossalai/deepseek_moe_7b_pretrain.sh
similarity index 100%
rename from nlp/llm/deepseek_moe_7b/ColossalAI/deepseek_moe_7b_pretrain.sh
rename to nlp/llm/deepseek_moe_7b/colossalai/deepseek_moe_7b_pretrain.sh
diff --git a/nlp/llm/deepseek_moe_7b/ColossalAI/model_utils.py b/nlp/llm/deepseek_moe_7b/colossalai/model_utils.py
similarity index 100%
rename from nlp/llm/deepseek_moe_7b/ColossalAI/model_utils.py
rename to nlp/llm/deepseek_moe_7b/colossalai/model_utils.py
diff --git a/nlp/llm/deepseek_moe_7b/ColossalAI/performance_evaluator.py b/nlp/llm/deepseek_moe_7b/colossalai/performance_evaluator.py
similarity index 100%
rename from nlp/llm/deepseek_moe_7b/ColossalAI/performance_evaluator.py
rename to nlp/llm/deepseek_moe_7b/colossalai/performance_evaluator.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/README.md b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/README.md
similarity index 45%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/README.md
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/README.md
index 26198825dbb49587192223f471bd4f2d77c19ca9..7fdca2d6b9c9d5b04955cbbed4351f8de11cc729 100644
--- a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/README.md
+++ b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/README.md
@@ -1,44 +1,45 @@
-# Llama3-8B (ColossalAI)
+# Llama3-8B SFT (ColossalAI)
 
 ## Model description
+
 The Llama 3 Herd of models natively supports multilinguality, coding, reasoning, and tool usage. Our largest model is dense Transformer with 405B parameters, processing information in a context window of up to 128K tokens, Llama 3 8B is the smallest model of Llama 3 Herd of models.
 
 ## Step 1: Installation
 
 Firstly, you should ensure that the corresponding version of ColossalAI has been installed in the iluvatar environment. Then install applications as follows:
 
-```bash
-$ cd ColossalAI/applications/Colossal-LLaMA
-$ pip3 install -e . 
+```sh
+cd ColossalAI/applications/Colossal-LLaMA
+pip3 install -e . 
 ```
 
 ## Step 2: Preparing datasets and checkpoints
 
-```bash
-$ pip3 install modelscope
-$ python3 ./get_Meta_LLaMA_8B.py
-$ mkdir -p /home/model_zoos/nlp
-$ mv ~/.cache/modelscope/hub/LLM-Research/Meta-Llama-3-8B /home/model_zoos/nlp
+```sh
+pip3 install modelscope
+python3 ./get_Meta_LLaMA_8B.py
+mkdir -p /home/model_zoos/nlp
+mv ~/.cache/modelscope/hub/LLM-Research/Meta-Llama-3-8B /home/model_zoos/nlp
 
-$ wget http://files.deepspark.org.cn:880/deepspark/tokenizer.model
-$ cp tokenizer.model /home/model_zoos/nlp/Meta-Llama-3-8B
+wget http://files.deepspark.org.cn:880/deepspark/tokenizer.model
+cp tokenizer.model /home/model_zoos/nlp/Meta-Llama-3-8B
 
-$ wget http://files.deepspark.org.cn:880/deepspark/school_math_0.25M.jsonl
-$ mv school_math_0.25M.jsonl dataset/school_math
-$ bash ./prepare_sft_dataset.sh llama3
+wget http://files.deepspark.org.cn:880/deepspark/school_math_0.25M.jsonl
+mv school_math_0.25M.jsonl dataset/school_math
+bash ./prepare_sft_dataset.sh llama3
 ```
 
 ## Step 3: Training
-```bash
-$ bash run_llama3_8b_sft_3d.sh
+
+```sh
+bash run_llama3_8b_sft_3d.sh
 ```
 
 ## Results
 
-| No.  | model     | peft        |    num_gpus        |train_samples_per_second |
-| ---- | --------- | ----------- | ------------------ | ----------------------  |
-| 1    | llama3-8b | Full sft    | 16                 |         1.53            |
-
+| model     | peft        |    num_gpus        |train_samples_per_second |
+| --------- | ----------- | ------------------ | ----------------------  |
+| llama3-8b | Full sft    | 16                 |         1.53            |
 
 ## Reference
 
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/__init__.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/__init__.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/__init__.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/__init__.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/dataset/__init__.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/dataset/__init__.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/dataset/__init__.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/dataset/__init__.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/dataset/conversation.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/dataset/conversation.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/dataset/conversation.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/dataset/conversation.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/dataset/dummy_dataset.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/dataset/dummy_dataset.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/dataset/dummy_dataset.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/dataset/dummy_dataset.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/dataset/loader.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/dataset/loader.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/dataset/loader.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/dataset/loader.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/dataset/spliced_and_tokenized_dataset.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/dataset/spliced_and_tokenized_dataset.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/dataset/spliced_and_tokenized_dataset.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/dataset/spliced_and_tokenized_dataset.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/model/init_model.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/model/init_model.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/model/init_model.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/model/init_model.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/tokenizer/init_tokenizer.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/tokenizer/init_tokenizer.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/tokenizer/init_tokenizer.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/tokenizer/init_tokenizer.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/utils/__init__.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/utils/__init__.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/utils/__init__.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/utils/__init__.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/utils/ckpt_io.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/utils/ckpt_io.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/utils/ckpt_io.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/utils/ckpt_io.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/utils/froze.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/utils/froze.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/utils/froze.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/utils/froze.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/utils/neftune_patch.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/utils/neftune_patch.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/utils/neftune_patch.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/utils/neftune_patch.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/utils/stream_chat_patch.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/utils/stream_chat_patch.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/utils/stream_chat_patch.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/utils/stream_chat_patch.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/utils/utils.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/utils/utils.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/colossal_llama/utils/utils.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/colossal_llama/utils/utils.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/dataset/convert_data.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/dataset/convert_data.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/dataset/convert_data.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/dataset/convert_data.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/dataset/prepare_pretrain_dataset.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/dataset/prepare_pretrain_dataset.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/dataset/prepare_pretrain_dataset.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/dataset/prepare_pretrain_dataset.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/dataset/prepare_sft_dataset.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/dataset/prepare_sft_dataset.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/dataset/prepare_sft_dataset.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/dataset/prepare_sft_dataset.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/get_Meta_LLaMA_8B.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/get_Meta_LLaMA_8B.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/get_Meta_LLaMA_8B.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/get_Meta_LLaMA_8B.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/performance_evaluator.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/performance_evaluator.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/performance_evaluator.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/performance_evaluator.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/prepare_sft_dataset.sh b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/prepare_sft_dataset.sh
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/prepare_sft_dataset.sh
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/prepare_sft_dataset.sh
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/requirements.txt b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/requirements.txt
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/requirements.txt
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/requirements.txt
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/run_llama3_8b_sft_3d.sh b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/run_llama3_8b_sft_3d.sh
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/run_llama3_8b_sft_3d.sh
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/run_llama3_8b_sft_3d.sh
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/setup.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/setup.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/setup.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/setup.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/train.example.sh b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/train.example.sh
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/train.example.sh
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/train.example.sh
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/train.py b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/train.py
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/train.py
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/train.py
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/train_sft.example.sh b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/train_sft.example.sh
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/train_sft.example.sh
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/train_sft.example.sh
diff --git a/nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/version.txt b/nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/version.txt
similarity index 100%
rename from nlp/llm/llama3_8b/ColossalAI/applications/Colossal-LLaMA/version.txt
rename to nlp/llm/llama3_8b/colossalai/applications/Colossal-LLaMA/version.txt
diff --git a/nlp/llm/llama3-8b/megatron-deepspeed/README.md b/nlp/llm/llama3_8b/megatron-deepspeed/README.md
similarity index 100%
rename from nlp/llm/llama3-8b/megatron-deepspeed/README.md
rename to nlp/llm/llama3_8b/megatron-deepspeed/README.md
diff --git a/nlp/llm/qwen2.5-7b/LLaMA-Factory/README.md b/nlp/llm/qwen2.5-7b/LLaMA-Factory/README.md
index 9331c4b86a3a72586fe8c0915d1ff574a025760c..65c87312ce9fca5ed298f721b39bcb8b7af027cb 100644
--- a/nlp/llm/qwen2.5-7b/LLaMA-Factory/README.md
+++ b/nlp/llm/qwen2.5-7b/LLaMA-Factory/README.md
@@ -1,4 +1,4 @@
-# Qwen2.5-7B
+# Qwen2.5-7B SFT (LLaMA-Factory)
 
 ## Model description