SOTAVerified

Natural Language Understanding

Natural Language Understanding is an important field of Natural Language Processing which contains various tasks such as text classification, natural language inference and story comprehension. Applications enabled by natural language understanding range from question answering to automated reasoning.

Source: Find a Reasonable Ending for Stories: Does Logic Relation Help the Story Cloze Test?

Papers

Showing 301350 of 1978 papers

TitleStatusHype
Supervised Contrastive Learning for Pre-trained Language Model Fine-tuningCode1
Comparison by Conversion: Reverse-Engineering UCCA from Syntax and Lexical SemanticsCode1
Adapting Pretrained Transformer to Lattices for Spoken Language UnderstandingCode1
ARBML: Democritizing Arabic Natural Language Processing ToolsCode1
RussianSuperGLUE: A Russian Language Understanding Evaluation BenchmarkCode1
FastFormers: Highly Efficient Transformer Models for Natural Language UnderstandingCode1
Towards Interpretable Natural Language Understanding with Explanations as Latent VariablesCode1
Pre-training Text-to-Text Transformers for Concept-centric Common SenseCode1
BARThez: a Skilled Pretrained French Sequence-to-Sequence ModelCode1
Optimal Subarchitecture Extraction For BERTCode1
RiSAWOZ: A Large-Scale Multi-Domain Wizard-of-Oz Dataset with Rich Semantic Annotations for Task-Oriented Dialogue ModelingCode1
Improving Constituency Parsing with Span AttentionCode1
Pretrained Language Models for Dialogue Generation with Multiple Input SourcesCode1
Incorporating BERT into Parallel Sequence Decoding with AdaptersCode1
TextHide: Tackling Data Privacy in Language Understanding TasksCode1
MultiWOZ 2.3: A multi-domain task-oriented dialogue dataset enhanced with annotation corrections and co-reference annotationCode1
COMET-ATOMIC 2020: On Symbolic and Neural Commonsense Knowledge GraphsCode1
Dual Inference for Improving Language Understanding and GenerationCode1
An Empirical Study of Tokenization Strategies for Various Korean NLP TasksCode1
Self-training Improves Pre-training for Natural Language UnderstandingCode1
Neural RST-based Evaluation of Discourse CoherenceCode1
A Simple but Tough-to-Beat Data Augmentation Approach for Natural Language Understanding and GenerationCode1
DialoGLUE: A Natural Language Understanding Benchmark for Task-Oriented DialogueCode1
Ape210K: A Large-Scale and Template-Rich Dataset of Math Word ProblemsCode1
Vector Projection Network for Few-shot Slot Tagging in Natural Language UnderstandingCode1
IndoNLU: Benchmark and Resources for Evaluating Indonesian Natural Language UnderstandingCode1
AutoTrans: Automating Transformer Design via Reinforced Architecture SearchCode1
VisualSem: A High-quality Knowledge Graph for Vision and LanguageCode1
ConvBERT: Improving BERT with Span-based Dynamic ConvolutionCode1
LogiQA: A Challenge Dataset for Machine Reading Comprehension with Logical ReasoningCode1
Towards Debiasing Sentence RepresentationsCode1
Evaluation Toolkit For Robustness Testing Of Automatic Essay Scoring SystemsCode1
Advances of Transformer-Based Models for News Headline GenerationCode1
tBERT: Topic Models and BERT Joining Forces for Semantic Similarity DetectionCode1
Rethinking Positional Encoding in Language Pre-trainingCode1
Multi-branch Attentive TransformerCode1
EPIE Dataset: A Corpus For Possible Idiomatic ExpressionsCode1
A Generative Model for Joint Natural Language Understanding and GenerationCode1
MC-BERT: Efficient Language Pre-Training via a Meta ControllerCode1
EPIC30M: An Epidemics Corpus Of Over 30 Million Relevant TweetsCode1
A Dataset for Statutory Reasoning in Tax Law Entailment and Question AnsweringCode1
State of the Art Models for Fake News Detection TasksCode1
A Top-Down Neural Architecture towards Text-Level Parsing of Discourse Rhetorical StructureCode1
Words aren't enough, their order matters: On the Robustness of Grounding Visual Referring ExpressionsCode1
IsoBN: Fine-Tuning BERT with Isotropic Batch NormalizationCode1
KLEJ: Comprehensive Benchmark for Polish Language UnderstandingCode1
Mind the Trade-off: Debiasing NLU Models without Degrading the In-distribution PerformanceCode1
TAVAT: Token-Aware Virtual Adversarial Training for Language UnderstandingCode1
Towards Unsupervised Language Understanding and Generation by Joint Dual LearningCode1
Lexical Semantic RecognitionCode1
Show:102550
← PrevPage 7 of 40Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1HNNAccuracy90Unverified
2BERT-large 340MAccuracy78.3Unverified
3UDSSM-II (ensemble)Accuracy78.3Unverified
4UDSSM-I (ensemble)Accuracy76.7Unverified
5DSSMAccuracy75Unverified
6UDSSM-IIAccuracy75Unverified
7BERT-base 110M + MASAccuracy68.3Unverified
8USSM + Supervised Deepnet + 3 Knowledge BasesAccuracy66.7Unverified
9Word-level CNN+LSTM (full scoring)Accuracy60Unverified
10Subword-level Transformer LMAccuracy58.3Unverified
#ModelMetricClaimedVerifiedStatus
1BERT (pred POS/lemmas)Tags (Full) Acc82.5Unverified
2BERT (none)Tags (Full) Acc82Unverified
3BERT (gold POS/lemmas)Tags (Full) Acc81Unverified
4GloVe (gold POS/lemmas)Tags (Full) Acc79.3Unverified
5RoBERTa + LinearFull F1 (Preps)78.2Unverified
6GloVe (none)Tags (Full) Acc77.5Unverified
7GloVe (pred POS/lemmas)Tags (Full) Acc77.1Unverified
8SVM (feature-rich, gold syntax)Role F1 (Preps)62.2Unverified
9BiLSTM + MLP (gold syntax)Role F1 (Preps)62.2Unverified
10SVM (feature-rich, auto syntax)Role F1 (Preps)58.2Unverified
#ModelMetricClaimedVerifiedStatus
1CaseLaw-BERTCaseHOLD75.6Unverified
2Legal-BERTCaseHOLD75.1Unverified
3DeBERTaCaseHOLD72.1Unverified
4LongformerCaseHOLD72Unverified
5RoBERTaCaseHOLD71.7Unverified
6BERTCaseHOLD70.7Unverified
7BigBirdCaseHOLD70.4Unverified
#ModelMetricClaimedVerifiedStatus
1ConvBERT-DGAverage74.6Unverified
2ConvBERT-DG + Pre + MultiAverage73.8Unverified
3mslmAverage73.49Unverified
4ConvBERT + Pre + MultiAverage68.22Unverified
5BanLanGenAverage39.16Unverified
#ModelMetricClaimedVerifiedStatus
1ConvBERT + Pre + MultiAverage86.89Unverified
2mslmAverage85.83Unverified
3ConvBERT-DG + Pre + MultiAverage85.34Unverified
#ModelMetricClaimedVerifiedStatus
1MT-DNN-SMARTAverage89.9Unverified
2BERT-LARGEAverage82.1Unverified