We demonstrate all pre-built applications in HugNLP.

Applications	Runing Tasks	Task Notes	PLM Models	Documents
Default Application	run_seq_cls.sh	Goal: Standard Fine-tuning or Prompt-tuning for sequence classification on user-defined dataset. Path: applications/default_applications	BERT, RoBERTa, DeBERTa	click
	run_seq_labeling.sh	Goal: Standard Fine-tuning for sequence labeling on user-defined dataset. Path: applications/default_applications	BERT, RoBERTa, ALBERT
Pre-training	run_pretrain_mlm.sh	Goal: Pre-training via Masked Language Modeling (MLM). Path: applications/pretraining/	BERT, RoBERTa	click
	run_pretrain_casual_lm.sh	Goal: Pre-training via Causal Language Modeling (CLM). Path: applications/pretraining	BERT, RoBERTa	click
GLUE Benchmark	run_glue.sh	Goal: Standard Fine-tuning or Prompt-tuning for GLUE classification tasks. Path: applications/benchmark/glue	BERT, RoBERTa, DeBERTa
	run_causal_incontext_glue.sh	Goal: In-context learning for GLUE classification tasks. Path: applications/benchmark/glue	GPT-2
CLUE Benchmark	clue_finetune_dev.sh	Goal: Standard Fine-tuning and Prompt-tuning for CLUE classification task。 Path: applications/benchmark/clue	BERT, RoBERTa, DeBERTa
	run_clue_cmrc.sh	Goal: Standard Fine-tuning for CLUE CMRC2018 task. Path: applications/benchmark/cluemrc	BERT, RoBERTa, DeBERTa
	run_clue_c3.sh	Goal: Standard Fine-tuning for CLUE C3 task. Path: applications/benchmark/cluemrc	BERT, RoBERTa, DeBERTa
	run_clue_chid.sh	Goal: Standard Fine-tuning for CLUE CHID task. Path: applications/benchmark/cluemrc	BERT, RoBERTa, DeBERTa
Instruction-Prompting	run_causal_instruction.sh	Goal: Cross-task training via generative Instruction-tuning based on causal PLM. You can use it to train a small ChatGPT. Path: applications/instruction_prompting/instruction_tuning	GPT2	click
	run_zh_extract_instruction.sh	Goal: Cross-task training via extractive Instruction-tuning based on Global Pointer model. Path: applications/instruction_prompting/chinese_instruction	BERT, RoBERTa, DeBERTa	click
	run_causal_incontext_cls.sh	Goal: In-context learning for user-defined classification tasks. Path: applications/instruction_prompting/incontext_learning	GPT-2	click
Information Extraction	run_extractive_unified_ie.sh	Goal: HugIE: training a unified chinese information extraction via extractive instruction-tuning. Path: applications/information_extraction/HugIE	BERT, RoBERTa, DeBERTa	click
	api_test.py	Goal: HugIE: API test. Path: applications/information_extraction/HugIE	-	click
	run_fewnerd.sh	Goal: Prototypical learning for named entity recognition, including SpanProto, TokenProto Path: applications/information_extraction/fewshot_ner	BERT
Code NLU	run_clone_cls.sh	Goal: Standard Fine-tuning for code clone classification task. Path: applications/code/code_clone	CodeBERT, CodeT5, GraphCodeBERT, PLBART	click
	run_defect_cls.sh	Goal: Standard Fine-tuning for code defect classification task. Path: applications/code/code_defect	CodeBERT, CodeT5, GraphCodeBERT, PLBART	click

Pre-built Application Settings

We show the settings that matched with each pre-built application.

Notes:

• ✅: Have finished
• ⌛️: To do
• ⛔️: Not-available

Applications	Runing Tasks	Adv-training	Parameter-efficient	Pattern-Verbalizer	Instruction-Prompting	Self-training	Calibration
Default Application	run_seq_cls.sh	✅	✅	✅
	run_seq_labeling.sh	✅	✅	✅
Pre-training	run_pretrain_mlm.sh	✅
	run_pretrain_casual_lm.sh	✅
GLUE Benchmark	run_glue.sh	✅	✅	✅
	run_causal_incontext_glue.sh	✅	✅	✅	✅		✅
CLUE Benchmark	clue_finetune_dev.sh	✅	✅	✅
	run_clue_cmrc.sh	✅
	run_clue_c3.sh	✅
	run_clue_chid.sh	✅
Instruction-Prompting	run_causal_instruction.sh	✅			✅
	run_zh_extract_instruction.sh	✅		✅	✅
	run_causal_incontext_cls.sh	⛔️	⛔️	✅	✅		✅
Information Extraction	run_extractive_unified_ie.sh	✅
	api_test.py	⛔️
	run_fewnerd.sh	✅
Code NLU	run_clone_cls.sh	✅	✅
	run_defect_cls.sh	✅	✅