SOTAVerified

Text Classification

Text Classification is the task of assigning a sentence or document an appropriate category. The categories depend on the chosen dataset and can range from topics.

Text Classification problems include emotion classification, news classification, citation intent classification, among others. Benchmark datasets for evaluating text classification capabilities include GLUE, AGNews, among others.

In recent years, deep learning techniques like XLNet and RoBERTa have attained some of the biggest performance jumps for text classification problems.

( Image credit: Text Classification Algorithms: A Survey )

Papers

Showing 151–200 of 3635 papers

TitleStatusHype
BertGCN: Transductive Text Classification by Combining GNN and BERTCode1
Better Robustness by More Coverage: Adversarial Training with Mixup Augmentation for Robust Fine-tuningCode1
Beyond the Next Token: Towards Prompt-Robust Zero-Shot Classification via Efficient Multi-Token PredictionCode1
Beyond the Tip of the Iceberg: Assessing Coherence of Text ClassifiersCode1
Black-Box Attacks on Sequential Recommenders via Data-Free Model ExtractionCode1
Black-box Generation of Adversarial Text Sequences to Evade Deep Learning ClassifiersCode1
Blacklight: Scalable Defense for Neural Networks against Query-Based Black-Box AttacksCode1
Building Efficient Universal Classifiers with Natural Language InferenceCode1
An Embarrassingly Simple Method to Mitigate Undesirable Properties of Pretrained Language Model TokenizersCode1
Challenges in Automated Debiasing for Toxic Language DetectionCode1
BanglaBERT: Language Model Pretraining and Benchmarks for Low-Resource Language Understanding Evaluation in BanglaCode1
ChatAgri: Exploring Potentials of ChatGPT on Cross-linguistic Agricultural Text ClassificationCode1
Balancing Methods for Multi-label Text Classification with Long-Tailed Class DistributionCode1
Class Adaptive Network CalibrationCode1
AcTune: Uncertainty-Based Active Self-Training for Active Fine-Tuning of Pretrained Language ModelsCode1
A Hybrid Classification Approach using Topic Modeling and Graph Convolution NetworksCode1
Closing the Loop: Fast, Interactive Semi-Supervised Annotation With Queries on Features and InstancesCode1
Cluster & Tune: Boost Cold Start Performance in Text ClassificationCode1
CODE-ACCORD: A Corpus of Building Regulatory Data for Rule Generation towards Automatic Compliance CheckingCode1
CogLTX: Applying BERT to Long TextsCode1
Compositional Generalization for Multi-label Text Classification: A Data-Augmentation ApproachCode1
Concept-Based Label Embedding via Dynamic Routing for Hierarchical Text ClassificationCode1
Concept Matching for Low-Resource ClassificationCode1
Consistency Regularization for Cross-Lingual Fine-TuningCode1
ContrastNet: A Contrastive Learning Framework for Few-Shot Text ClassificationCode1
ContrXT: Generating Contrastive Explanations from any Text ClassifierCode1
Cost-Sensitive BERT for Generalisable Sentence Classification with Imbalanced DataCode1
Counterfactual Inference for Text Classification DebiasingCode1
Behavioral Factors in Interactive Training of Text ClassifiersCode1
CRUISE-Screening: Living Literature Reviews ToolboxCode1
Bag of Tricks for Efficient Text ClassificationCode1
Cyclical Curriculum LearningCode1
Are We Really Making Much Progress in Text Classification? A Comparative ReviewCode1
Be More with Less: Hypergraph Attention Networks for Inductive Text ClassificationCode1
AlleNoise: large-scale text classification benchmark dataset with real-world label noiseCode1
Denoising Multi-Source Weak Supervision for Neural Text ClassificationCode1
ALL-IN-1: Short Text Classification with One Model for All LanguagesCode1
Diffuser: Efficient Transformers with Multi-hop Attention Diffusion for Long SequencesCode1
DisCo: Distilled Student Models Co-training for Semi-supervised Text MiningCode1
DiscoPrompt: Path Prediction Prompt Tuning for Implicit Discourse Relation RecognitionCode1
CARE: Collaborative AI-Assisted Reading EnvironmentCode1
An Empirical Study of Invariant Risk MinimizationCode1
DKEC: Domain Knowledge Enhanced Multi-Label Classification for Diagnosis PredictionCode1
DocBERT: BERT for Document ClassificationCode1
Domain-Adaptive Text Classification with Structured Knowledge from Unlabeled DataCode1
Don’t Miss the Labels: Label-semantic Augmented Meta-Learner for Few-Shot Text ClassificationCode1
Automated Essay Scoring Using Transformer ModelsCode1
Dual Contrastive Learning: Text Classification via Label-Aware Data AugmentationCode1
Active Learning for BERT: An Empirical StudyCode1
BootAug: Boosting Text Augmentation via Hybrid Instance Filtering FrameworkCode1
Show:102550
← PrevPage 4 of 73Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ST5-XXLAccuracy73.42—Unverified
2ST5-XLAccuracy72.84—Unverified
3ST5-LargeAccuracy72.31—Unverified
4Ada SimilarityAccuracy70.44—Unverified
5SGPT-5.8B-nliAccuracy70.14—Unverified
6ST5-BaseAccuracy69.81—Unverified
7SGPT-5.8B-msmarcoAccuracy68.13—Unverified
8MPNet-multilingualAccuracy67.91—Unverified
9GTR-XXLAccuracy67.41—Unverified
10SimCSE-BERT-supAccuracy67.32—Unverified
#ModelMetricClaimedVerifiedStatus
1Mistral-Small-24B + CAPOError15.7—Unverified
2ToWE-SGError14—Unverified
3Qwen2.5-32B + CAPOError12.93—Unverified
4Llama-3.3-70B + CAPOError11.2—Unverified
5Seq2CNN with GWS(50)Error9.64—Unverified
6Char-level CNNError9.51—Unverified
7SVDCNNError9.45—Unverified
8VDCNError8.67—Unverified
9Balanced+bi-leaf-RNNError7.9—Unverified
10CCCapsNetError7.61—Unverified
#ModelMetricClaimedVerifiedStatus
1Seq2CNN(50)Error2.77—Unverified
2Char-level CNNError1.55—Unverified
3SWEM-concatError1.43—Unverified
4FastTextError1.4—Unverified
5VDCNError1.29—Unverified
6CCCapsNetError1.28—Unverified
7Balanced+bi-leaf-RNNError1.2—Unverified
8BERT large UDAError1.09—Unverified
9M-ACNNError1.07—Unverified
10EXAMError1—Unverified
#ModelMetricClaimedVerifiedStatus
1DeBERTaAccuracy98.45—Unverified
2C-BERT (ESGNN + BERT)Accuracy98.28—Unverified
3ESGNNAccuracy98.23—Unverified
4RoBERTaGCNAccuracy98.2—Unverified
5BERTAccuracy98.17—Unverified
6SGNNAccuracy98.09—Unverified
7ERNIE 2.0Accuracy98.04—Unverified
8DistilBERTAccuracy97.98—Unverified
9Our Model*Accuracy97.8—Unverified
10ALBERTv2Accuracy97.62—Unverified
#ModelMetricClaimedVerifiedStatus
1TM-GloveError9.96—Unverified
2byte mLSTM7Error9.6—Unverified
3DELTA (CNN)Error7.8—Unverified
4SWEM-averError7.8—Unverified
5Capsule-BError7.2—Unverified
6STM+TSED+PT+2LError7.04—Unverified
7GRU-RNN-GLOVEError7—Unverified
8MPAD-pathError6.2—Unverified
9VLAWEError5.8—Unverified
10C-LSTMError5.4—Unverified
#ModelMetricClaimedVerifiedStatus
1LinearSVM+TFIDFAccuracy93—Unverified
2RoBERTaGCNAccuracy89.5—Unverified
3SSGCAccuracy88.6—Unverified
4SGCAccuracy88.5—Unverified
5SGCNAccuracy88.5—Unverified
6RMDL (15 RDLs)Accuracy87.91—Unverified
7Sparse Tensor ClassifierAccuracy87.3—Unverified
8GraphStarAccuracy86.9—Unverified
9NABoE-fullAccuracy86.8—Unverified
10Text GCNAccuracy86.34—Unverified
#ModelMetricClaimedVerifiedStatus
1ELECTRA + ANNF199.6—Unverified
2ERNIE + ANNF199.4—Unverified
3XLNet + ANNF199.2—Unverified
4RoBERTa + ANNF198.7—Unverified
5Longformer + ANNF193.9—Unverified
6BERT + ANNF190.5—Unverified
7ALBERT + ANNF179.7—Unverified
8BERTF175—Unverified
9DistilBERTF174.4—Unverified
10LongformerF174—Unverified
#ModelMetricClaimedVerifiedStatus
1RoBERTaGCNAccuracy72.8—Unverified
2Our Model*Accuracy69.4—Unverified
3SGCNAccuracy68.5—Unverified
4SGCAccuracy68.5—Unverified
5SSGCAccuracy68.5—Unverified
6Text GCNAccuracy68.36—Unverified
7GraphStarAccuracy64.2—Unverified
8ApproxRepSetAccuracy64.06—Unverified
9REL-RWMD k-NNAccuracy58.74—Unverified
10CNN+LowercasedAccuracy36.2—Unverified
#ModelMetricClaimedVerifiedStatus
1BERT-ITPT-FiTAccuracy77.62—Unverified
2DRNNAccuracy76.26—Unverified
3DELTA (HAN)Accuracy75.1—Unverified
4EXAMAccuracy74.8—Unverified
5DNC+CUWAccuracy74.3—Unverified
6ULMFiT (Small data)Accuracy74.3—Unverified
7CCCapsNetAccuracy73.85—Unverified
8SWEM-concatAccuracy73.53—Unverified
9FastTextAccuracy72.3—Unverified
10Seq2CNN(50)Accuracy55.39—Unverified
#ModelMetricClaimedVerifiedStatus
1DeBERTaAccuracy90.21—Unverified
2RoBERTaGCNAccuracy89.7—Unverified
3ERNIE 2.0 (optimized)Accuracy89.53—Unverified
4RoBERTaAccuracy89.42—Unverified
5ERNIE 2.0Accuracy88.97—Unverified
6BERTAccuracy86.94—Unverified
7ALBERTv2Accuracy86.02—Unverified
8DistilBERTAccuracy85.31—Unverified
9SSGCAccuracy76.7—Unverified
#ModelMetricClaimedVerifiedStatus
1CliReBERT (P0L3/clirebert_clirevocab_uncased)Evaluation Macro F10.65—Unverified
2ClimateBERT (climatebert/distilroberta-base-climate-f)Evaluation Macro F10.64—Unverified
3BERT (google-bert/bert-base-uncased)Evaluation Macro F10.61—Unverified
4CliSciBERT (P0L3/cliscibert_scivocab_uncased)Evaluation Macro F10.61—Unverified
5SciBERT (allenai/scibert_scivocab_cased)Evaluation Macro F10.59—Unverified
6DistilRoBERTa (distilbert/distilroberta-base)Evaluation Macro F10.58—Unverified
7SciClimateBERT (P0L3/sciclimatebert)Evaluation Macro F10.58—Unverified
8RoBERTa (FacebookAI/roberta-base)Evaluation Macro F10.57—Unverified
#ModelMetricClaimedVerifiedStatus
1Human (Post-Rec.) (Spangher et al., 2021)macro F173.69—Unverified
2MT-Mac (Spangher et al., 2021)macro F163.46—Unverified
3MT-Mic (Spangher et al., 2021)macro F161.89—Unverified
4RL-IP/TT (Choubey et al., 2021)macro F157—Unverified
5Document LSTM + Document encoding (Choubey et al., 2020)macro F154.4—Unverified
6CRF Fine-grained (Choubey et al., 2020)macro F152.9—Unverified
7Human (Blind) (Spangher et al., 2021)macro F146.18—Unverified
8Feature-based (SVM) (Choubey et al., 2020)macro F138.3—Unverified
#ModelMetricClaimedVerifiedStatus
11-6 BertGCNAccuracy96.6—Unverified
2GraphStarAccuracy95—Unverified
3Our Model*Accuracy94.6—Unverified
4SSGCAccuracy94.5—Unverified
5SGCNAccuracy94—Unverified
6SGCAccuracy94—Unverified
7Text GCNAccuracy93.56—Unverified
8TM-GloveAccuracy89.14—Unverified