Topic Classification on NYT (test)
92.5AccuracyNLI-ST
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| NLI-STLearning Paradigm=Reference, Use auxiliary labeled data=true, Use task-specific corpus=true2023.05 | 92.5 | — | — | |
| GoldTraining Set=Original training set2023.06 | 83.8 | 81.02 | — | |
| AttrPromptTraining Set=Synthetic data created with AttrPrompt2023.06 | 81.3 | 82.26 | 1.05 | |
| AttrPrompt w/o CAFTraining Set=Synthetic data created with AttrPrompt without CAF2023.06 | 80.4 | 80.92 | 0.91 | |
| MetaPromptTraining Set=Synthetic data created with MetaPrompt2023.06 | 79.58 | 79.83 | 0.87 | |
| X-ClassLearning Paradigm=Reference, Use task-specific corpus=true2023.05 | 78.8 | — | — | |
| SimPromptTraining Set=Synthetic data created with SimPrompt2023.06 | 75.47 | 76.22 | 0.76 | |
| REGENLearning Paradigm=Zero-shot Learning via Generating Task-specific Datasets, Standard Deviation=1.12023.05 | 74.5 | — | — | |
| LLM Zero-ShotMode=Zero-shot predictor2023.06 | 74.16 | 69.84 | 5.44 | |
| KPTLearning Paradigm=Reference, Use task-specific corpus=true, Use additional knowledge base=true2023.05 | 72.1 | — | — | |
| TE-NLI (Best)Learning Paradigm=Labeled data usage, Use auxiliary labeled data=true2023.05 | 70.7 | — | — | |
| Mining (Re-implementation)Learning Paradigm=Zero-shot Learning via Generating Task-specific Datasets, Concurrent Work=true, Fair Comparison Setup=true, Standard Deviation=0.92023.05 | 68.6 | — | — | |
| PromptLearning Paradigm=Zero-shot Learning via Direct Inferencing2023.05 | 57.4 | — | — | |
| GPT-3Learning Paradigm=Zero-shot Learning via Direct Inferencing, Billion-scale PLM=true2023.05 | 57 | — | — | |
| MiningLearning Paradigm=Zero-shot Learning via Generating Task-specific Datasets, Concurrent Work=true2023.05 | 56.1 | — | — | |
| NSP-BERTLearning Paradigm=Zero-shot Learning via Direct Inferencing2023.05 | 54.6 | — | — | |
| SuperGenLearning Paradigm=Zero-shot Learning via Generating Task-specific Datasets, Standard Deviation=1.52023.05 | 53.9 | — | — | |
| LOTClassLearning Paradigm=Reference, Use task-specific corpus=true2023.05 | 49.5 | — | — |