Merge branch 'main' into release/2.0

Author: tastelikefeet

README.md

@@ -523,20 +523,21 @@ The complete list of supported models and datasets can be found at [Supported Mo
 
 ### Supported Technologies
 
-| Technology Name                                               |
-|--------------------------------------------------------------- |
+| Technology Name                                              |
+| ------------------------------------------------------------ |
 | 🔥LoRA: [LORA: LOW-RANK ADAPTATION OF LARGE LANGUAGE MODELS](https://arxiv.org/abs/2106.09685) |
 | 🔥LoRA+: [LoRA+: Efficient Low Rank Adaptation of Large Models](https://arxiv.org/pdf/2402.12354.pdf) |
+| 🔥GaLore:[GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection](https://arxiv.org/abs/2403.03507) |
+| 🔥LISA: [LISA: Layerwise Importance Sampling for Memory-Efficient Large Language Model Fine-Tuning](https://arxiv.org/abs/2403.17919) |
+| 🔥UnSloth: https://github.com/unslothai/unsloth               |
 | 🔥LLaMA PRO: [LLAMA PRO: Progressive LLaMA with Block Expansion](https://arxiv.org/pdf/2401.02415.pdf) |
-| 🔥SCEdit: [SCEdit: Efficient and Controllable Image Diffusion Generation via Skip Connection Editing](https://arxiv.org/abs/2312.11392)  < [arXiv](https://arxiv.org/abs/2312.11392)  \|  [Project Page](https://scedit.github.io/) > |
+| 🔥SCEdit: [SCEdit: Efficient and Controllable Image Diffusion Generation via Skip Connection Editing](https://arxiv.org/abs/2312.11392)  < [arXiv](https://arxiv.org/abs/2312.11392)  \ |
 | 🔥NEFTune: [Noisy Embeddings Improve Instruction Finetuning](https://arxiv.org/abs/2310.05914) |
-| QA-LoRA:[Quantization-Aware Low-Rank Adaptation of Large Language Models](https://arxiv.org/abs/2309.14717) |
 | LongLoRA: [Efficient Fine-tuning of Long-Context Large Language Models](https://arxiv.org/abs/2309.12307) |
-| ROME: [Rank-One Editing of Encoder-Decoder Models](https://arxiv.org/abs/2211.13317) |
 | Adapter: [Parameter-Efficient Transfer Learning for NLP](http://arxiv.org/abs/1902.00751) |
-| Prompt Tuning: [Visual Prompt Tuning](https://arxiv.org/abs/2203.12119) |
+| Vision Prompt Tuning: [Visual Prompt Tuning](https://arxiv.org/abs/2203.12119) |
 | Side: [Side-Tuning: A Baseline for Network Adaptation via Additive Side Networks](https://arxiv.org/abs/1912.13503) |
-| Res-Tuning: [Res-Tuning: A Flexible and Efficient Tuning Paradigm via Unbinding Tuner from Backbone](https://arxiv.org/abs/2310.19859)  < [arXiv](https://arxiv.org/abs/2310.19859)  \|  [Project Page](https://res-tuning.github.io/)  \|  [Usage](docs/source/GetStarted/ResTuning.md) > |
+| Res-Tuning: [Res-Tuning: A Flexible and Efficient Tuning Paradigm via Unbinding Tuner from Backbone](https://arxiv.org/abs/2310.19859)  < [arXiv](https://arxiv.org/abs/2310.19859)  \ |
 | Tuners provided by [PEFT](https://github.com/huggingface/peft), such as IA3, AdaLoRA, etc. |
 
 ### Supported Hardware

README_CN.md

@@ -522,21 +522,22 @@ CUDA_VISIBLE_DEVICES=0 swift deploy \
 
 ### 支持的技术
 
-| 技术名称                                                     |
-| ------------------------------------------------------------ |
-| 🔥LoRA: [LORA: LOW-RANK ADAPTATION OF LARGE LANGUAGE MODELS](https://arxiv.org/abs/2106.09685) |
-| 🔥LoRA+: [LoRA+: Efficient Low Rank Adaptation of Large Models](https://arxiv.org/pdf/2402.12354.pdf) |
-| 🔥LLaMA PRO: [LLAMA PRO: Progressive LLaMA with Block Expansion](https://arxiv.org/pdf/2401.02415.pdf) |
-| 🔥SCEdit: [SCEdit: Efficient and Controllable Image Diffusion Generation via Skip Connection Editing](https://arxiv.org/abs/2312.11392)  < [arXiv](https://arxiv.org/abs/2312.11392)  \|  [Project Page](https://scedit.github.io/) > |
-| 🔥NEFTune: [Noisy Embeddings Improve Instruction Finetuning](https://arxiv.org/abs/2310.05914) |
-| QA-LoRA:[Quantization-Aware Low-Rank Adaptation of Large Language Models](https://arxiv.org/abs/2309.14717) |
-| LongLoRA: [Efficient Fine-tuning of Long-Context Large Language Models](https://arxiv.org/abs/2309.12307) |
-| ROME: [Rank-One Editing of Encoder-Decoder Models](https://arxiv.org/abs/2211.13317) |
-| Adapter: [Parameter-Efficient Transfer Learning for NLP](http://arxiv.org/abs/1902.00751) |
-| Prompt Tuning: [Visual Prompt Tuning](https://arxiv.org/abs/2203.12119) |
-| Side: [Side-Tuning: A Baseline for Network Adaptation via Additive Side Networks](https://arxiv.org/abs/1912.13503) |
-| Res-Tuning: [Res-Tuning: A Flexible and Efficient Tuning Paradigm via Unbinding Tuner from Backbone](https://arxiv.org/abs/2310.19859)  < [arXiv](https://arxiv.org/abs/2310.19859)  \|  [Project Page](https://res-tuning.github.io/)  \|  [Usage](docs/source/GetStarted/ResTuning.md) > |
-| [PEFT](https://github.com/huggingface/peft)提供的tuners, 如IA3, AdaLoRA等 |
+| 技术名称                                                                                                                                                                                    |
+|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| 🔥LoRA: [LORA: LOW-RANK ADAPTATION OF LARGE LANGUAGE MODELS](https://arxiv.org/abs/2106.09685)                                                                                          |
+| 🔥LoRA+: [LoRA+: Efficient Low Rank Adaptation of Large Models](https://arxiv.org/pdf/2402.12354.pdf)                                                                                   |
+| 🔥LLaMA PRO: [LLAMA PRO: Progressive LLaMA with Block Expansion](https://arxiv.org/pdf/2401.02415.pdf)                                                                                  |
+| 🔥GaLore:[GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection](https://arxiv.org/abs/2403.03507)                                                                      |
+| 🔥LISA: [LISA: Layerwise Importance Sampling for Memory-Efficient Large Language Model Fine-Tuning](https://arxiv.org/abs/2403.17919)                                                   |
+| 🔥UnSloth: https://github.com/unslothai/unsloth                                                                                                                                         |
+| 🔥SCEdit: [SCEdit: Efficient and Controllable Image Diffusion Generation via Skip Connection Editing](https://arxiv.org/abs/2312.11392)  < [arXiv](https://arxiv.org/abs/2312.11392)  \ |  [Project Page](https://scedit.github.io/) > |
+| 🔥NEFTune: [Noisy Embeddings Improve Instruction Finetuning](https://arxiv.org/abs/2310.05914)                                                                                          |
+| LongLoRA: [Efficient Fine-tuning of Long-Context Large Language Models](https://arxiv.org/abs/2309.12307)                                                                               |
+| Adapter: [Parameter-Efficient Transfer Learning for NLP](http://arxiv.org/abs/1902.00751)                                                                                               |
+| Vision Prompt Tuning: [Visual Prompt Tuning](https://arxiv.org/abs/2203.12119)                                                                                                          |
+| Side: [Side-Tuning: A Baseline for Network Adaptation via Additive Side Networks](https://arxiv.org/abs/1912.13503)                                                                     |
+| Res-Tuning: [Res-Tuning: A Flexible and Efficient Tuning Paradigm via Unbinding Tuner from Backbone](https://arxiv.org/abs/2310.19859)  < [arXiv](https://arxiv.org/abs/2310.19859)  \  |  [Project Page](https://res-tuning.github.io/)  \|  [Usage](docs/source/GetStarted/ResTuning.md) > |
+| [PEFT](https://github.com/huggingface/peft)提供的tuners, 如IA3, AdaLoRA等                                                                                                                    |
 
 ### 支持的硬件
 

docs/source/GetStarted/使用tuners.md

@@ -5,27 +5,28 @@ tuner是指附加在模型上的额外结构部分，用于减少训练参数量
 1. LoRA: [LORA: LOW-RANK ADAPTATION OF LARGE LANGUAGE MODELS](https://arxiv.org/abs/2106.09685)
 2. LoRA+: [LoRA+: Efficient Low Rank Adaptation of Large Models](https://arxiv.org/pdf/2402.12354.pdf)
 3. LLaMA PRO: [LLAMA PRO: Progressive LLaMA with Block Expansion](https://arxiv.org/pdf/2401.02415.pdf)
-4. SCEdit: [SCEdit: Efficient and Controllable Image Diffusion Generation via Skip Connection Editing](https://arxiv.org/abs/2312.11392)  < [arXiv](https://arxiv.org/abs/2312.11392)  |  [Project Page](https://scedit.github.io/) >
-5. NEFTune: [Noisy Embeddings Improve Instruction Finetuning](https://arxiv.org/abs/2310.05914)
-6. QA-LoRA:[Quantization-Aware Low-Rank Adaptation of Large Language Models](https://arxiv.org/abs/2309.14717).
-7. LongLoRA: [Efficient Fine-tuning of Long-Context Large Language Models](https://arxiv.org/abs/2309.12307)
-8. ROME: [Rank-One Editing of Encoder-Decoder Models](https://arxiv.org/abs/2211.13317)
-9. Adapter: [Parameter-Efficient Transfer Learning for NLP](http://arxiv.org/abs/1902.00751)
-10. Prompt Tuning: [Visual Prompt Tuning](https://arxiv.org/abs/2203.12119)
-11. Side: [Side-Tuning: A Baseline for Network Adaptation via Additive Side Networks](https://arxiv.org/abs/1912.13503)
-12. Res-Tuning: [Res-Tuning: A Flexible and Efficient Tuning Paradigm via Unbinding Tuner from Backbone](https://arxiv.org/abs/2310.19859)  < [arXiv](https://arxiv.org/abs/2310.19859)  |  [Project Page](https://res-tuning.github.io/)  |  [Usage](docs/source/GetStarted/ResTuning.md) >
-13. [PEFT](https://github.com/huggingface/peft)提供的tuners, 如IA3, AdaLoRA等
+4. GaLore: [GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection](https://arxiv.org/abs/2403.03507)
+5. LISA: [LISA: Layerwise Importance Sampling for Memory-Efficient Large Language Model Fine-Tuning](https://arxiv.org/abs/2403.17919)
+6. UnSloth: https://github.com/unslothai/unsloth
+7. SCEdit: [SCEdit: Efficient and Controllable Image Diffusion Generation via Skip Connection Editing](https://arxiv.org/abs/2312.11392)  < [arXiv](https://arxiv.org/abs/2312.11392)  |  [Project Page](https://scedit.github.io/) >
+8. NEFTune: [Noisy Embeddings Improve Instruction Finetuning](https://arxiv.org/abs/2310.05914)
+9. LongLoRA: [Efficient Fine-tuning of Long-Context Large Language Models](https://arxiv.org/abs/2309.12307)
+10. Adapter: [Parameter-Efficient Transfer Learning for NLP](http://arxiv.org/abs/1902.00751)
+11. Vision Prompt Tuning: [Visual Prompt Tuning](https://arxiv.org/abs/2203.12119)
+12. Side: [Side-Tuning: A Baseline for Network Adaptation via Additive Side Networks](https://arxiv.org/abs/1912.13503)
+13. Res-Tuning: [Res-Tuning: A Flexible and Efficient Tuning Paradigm via Unbinding Tuner from Backbone](https://arxiv.org/abs/2310.19859)  < [arXiv](https://arxiv.org/abs/2310.19859)  |  [Project Page](https://res-tuning.github.io/)  |  [Usage](docs/source/GetStarted/ResTuning.md) >
+14. [PEFT](https://github.com/huggingface/peft)提供的tuners, 如IA3, AdaLoRA等
 
 ## 在训练中使用
 
 调用`Swift.prepare_model()`来将tuners添加到模型上：
 
 ```python
 from modelscope import Model
-from swift import Swift, LoRAConfig
+from swift import Swift, LoraConfig
 import torch
 model = Model.from_pretrained('ZhipuAI/chatglm3-6b', torch_dtype=torch.bfloat16, device_map='auto')
-lora_config = LoRAConfig(
+lora_config = LoraConfig(
                 r=16,
                 target_modules=['query_key_value'],
                 lora_alpha=32,
@@ -37,10 +38,10 @@ model = Swift.prepare_model(model, lora_config)
 
 ```python
 from modelscope import Model
-from swift import Swift, LoRAConfig, AdapterConfig
+from swift import Swift, LoraConfig, AdapterConfig
 import torch
 model = Model.from_pretrained('ZhipuAI/chatglm3-6b', torch_dtype=torch.bfloat16, device_map='auto')
-lora_config = LoRAConfig(
+lora_config = LoraConfig(
                 r=16,
                 target_modules=['query_key_value'],
                 lora_alpha=32,
@@ -105,13 +106,13 @@ model.save_pretrained(save_directory='./output')
 from swift import Seq2SeqTrainer, Seq2SeqTrainingArguments
 from modelscope import MsDataset, AutoTokenizer
 from modelscope import AutoModelForCausalLM
-from swift import Swift, LoRAConfig
+from swift import Swift, LoraConfig
 from swift.llm import get_template, TemplateType
 import torch
 
 # 拉起模型
 model = AutoModelForCausalLM.from_pretrained('ZhipuAI/chatglm3-6b', torch_dtype=torch.bfloat16, device_map='auto', trust_remote_code=True)
-lora_config = LoRAConfig(
+lora_config = LoraConfig(
                 r=16,
                 target_modules=['query_key_value'],
                 lora_alpha=32,

docs/source_en/GetStarted/Tuners.md

@@ -5,27 +5,28 @@
 1. LoRA: [LORA: LOW-RANK ADAPTATION OF LARGE LANGUAGE MODELS](https://arxiv.org/abs/2106.09685)
 2. LoRA+: [LoRA+: Efficient Low Rank Adaptation of Large Models](https://arxiv.org/pdf/2402.12354.pdf)
 3. LLaMA PRO: [LLAMA PRO: Progressive LLaMA with Block Expansion](https://arxiv.org/pdf/2401.02415.pdf)
-4. SCEdit: [SCEdit: Efficient and Controllable Image Diffusion Generation via Skip Connection Editing](https://arxiv.org/abs/2312.11392)  < [arXiv](https://arxiv.org/abs/2312.11392)  |  [Project Page](https://scedit.github.io/) >
-5. NEFTune: [Noisy Embeddings Improve Instruction Finetuning](https://arxiv.org/abs/2310.05914)
-6. QA-LoRA:[Quantization-Aware Low-Rank Adaptation of Large Language Models](https://arxiv.org/abs/2309.14717).
-7. LongLoRA: [Efficient Fine-tuning of Long-Context Large Language Models](https://arxiv.org/abs/2309.12307)
-8. ROME: [Rank-One Editing of Encoder-Decoder Models](https://arxiv.org/abs/2211.13317)
-9. Adapter: [Parameter-Efficient Transfer Learning for NLP](http://arxiv.org/abs/1902.00751)
-10. Prompt Tuning: [Visual Prompt Tuning](https://arxiv.org/abs/2203.12119)
-11. Side: [Side-Tuning: A Baseline for Network Adaptation via Additive Side Networks](https://arxiv.org/abs/1912.13503)
-12. Res-Tuning: [Res-Tuning: A Flexible and Efficient Tuning Paradigm via Unbinding Tuner from Backbone](https://arxiv.org/abs/2310.19859)  < [arXiv](https://arxiv.org/abs/2310.19859)  |  [Project Page](https://res-tuning.github.io/)  |  [Usage](docs/source/GetStarted/ResTuning.md) >
-13. Tuners provided by [PEFT](https://github.com/huggingface/peft), such as IA3, AdaLoRA, etc.
+4. GaLore: [GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection](https://arxiv.org/abs/2403.03507)
+5. LISA: [LISA: Layerwise Importance Sampling for Memory-Efficient Large Language Model Fine-Tuning](https://arxiv.org/abs/2403.17919)
+6. UnSloth: https://github.com/unslothai/unsloth
+7. SCEdit: [SCEdit: Efficient and Controllable Image Diffusion Generation via Skip Connection Editing](https://arxiv.org/abs/2312.11392)  < [arXiv](https://arxiv.org/abs/2312.11392)  |  [Project Page](https://scedit.github.io/) >
+8. NEFTune: [Noisy Embeddings Improve Instruction Finetuning](https://arxiv.org/abs/2310.05914)
+9. LongLoRA: [Efficient Fine-tuning of Long-Context Large Language Models](https://arxiv.org/abs/2309.12307)
+10. Adapter: [Parameter-Efficient Transfer Learning for NLP](http://arxiv.org/abs/1902.00751)
+11. Vision Prompt Tuning: [Visual Prompt Tuning](https://arxiv.org/abs/2203.12119)
+12. Side: [Side-Tuning: A Baseline for Network Adaptation via Additive Side Networks](https://arxiv.org/abs/1912.13503)
+13. Res-Tuning: [Res-Tuning: A Flexible and Efficient Tuning Paradigm via Unbinding Tuner from Backbone](https://arxiv.org/abs/2310.19859)  < [arXiv](https://arxiv.org/abs/2310.19859)  |  [Project Page](https://res-tuning.github.io/)  |  [Usage](docs/source/GetStarted/ResTuning.md) >
+14. Tuners provided by [PEFT](https://github.com/huggingface/peft), such as IA3, AdaLoRA, etc.
 
 ## Using in Training
 
 Call `Swift.prepare_model()` to add tuners to the model:
 
 ```python
 from modelscope import Model
-from swift import Swift, LoRAConfig
+from swift import Swift, LoraConfig
 import torch
 model = Model.from_pretrained('ZhipuAI/chatglm3-6b', torch_dtype=torch.bfloat16, device_map='auto')
-lora_config = LoRAConfig(
+lora_config = LoraConfig(
                 r=16,
                 target_modules=['query_key_value'],
                 lora_alpha=32,
@@ -37,10 +38,10 @@ Multiple tuners can also be used simultaneously:
 
 ```python
 from modelscope import Model
-from swift import Swift, LoRAConfig, AdapterConfig
+from swift import Swift, LoraConfig, AdapterConfig
 import torch
 model = Model.from_pretrained('ZhipuAI/chatglm3-6b', torch_dtype=torch.bfloat16, device_map='auto')
-lora_config = LoRAConfig(
+lora_config = LoraConfig(
                 r=16,
                 target_modules=['query_key_value'],
                 lora_alpha=32,
@@ -105,13 +106,13 @@ If only a single config is passed in, the default name `default` will be used:
 from swift import Seq2SeqTrainer, Seq2SeqTrainingArguments
 from modelscope import MsDataset, AutoTokenizer
 from modelscope import AutoModelForCausalLM
-from swift import Swift, LoRAConfig
+from swift import Swift, LoraConfig
 from swift.llm import get_template, TemplateType
 import torch
 
 # load model
 model = AutoModelForCausalLM.from_pretrained('ZhipuAI/chatglm3-6b', torch_dtype=torch.bfloat16, device_map='auto', trust_remote_code=True)
-lora_config = LoRAConfig(
+lora_config = LoraConfig(
                 r=16,
                 target_modules=['query_key_value'],
                 lora_alpha=32,