ML Research Wiki / Benchmarks / Natural Language Inference / SNLI

SNLI

Natural Language Inference Benchmark

Performance Over Time

📊 Showing 88 results | 📏 Metric: % Test Accuracy

Top Performing Models

Rank	Model	Paper	% Test Accuracy	Date	Code
1	UnitedSynT5 (3B) 📚	First Train to Generate, then Generate to Train: UnitedSynT5 for Few-Shot NLI	94.70	2024-12-12	-
2	UnitedSynT5 (335M) 📚	First Train to Generate, then Generate to Train: UnitedSynT5 for Few-Shot NLI	93.50	2024-12-12	-
3	Neural Tree Indexers for Text Understanding	Entailment as Few-Shot Learner	93.10	2021-04-29	📦 PaddlePaddle/PaddleNLP 📦 sunyilgdx/prompts4keras 📦 cactilab/hateguard
4	EFL (Entailment as Few-shot Learner) + RoBERTa-large	Entailment as Few-Shot Learner	93.10	2021-04-29	📦 PaddlePaddle/PaddleNLP 📦 sunyilgdx/prompts4keras 📦 cactilab/hateguard
5	MT-DNN-SMARTLARGEv0	SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization	92.60	2019-11-08	📦 namisan/mt-dnn 📦 microsoft/MT-DNN 📦 archinetai/smart-pytorch
6	RoBERTa-large+Self-Explaining	Self-Explaining Structures Improve NLP Models	92.30	2020-12-03	📦 ShannonAI/Self_Explaining_Structures_Improve_NLP_Models
7	RoBERTa-large + self-explaining layer	Self-Explaining Structures Improve NLP Models	92.30	2020-12-03	📦 ShannonAI/Self_Explaining_Structures_Improve_NLP_Models
8	CA-MTL	Conditionally Adaptive Multi-Task Learning: Improving Transfer Learning in NLP Using Fewer Parameters & Less Data	92.10	2020-09-19	📦 CAMTL/CA-MTL
9	SemBERT	Semantics-aware BERT for Language Understanding	91.90	2019-09-05	📦 cooelf/SemBERT
10	MT-DNN	Multi-Task Deep Neural Networks for Natural Language Understanding	91.60	2019-01-31	📦 namisan/mt-dnn 📦 xycforgithub/MultiTask-MRC 📦 ABaldrati/MT-BERT

All Papers (88)

First Train to Generate, then Generate to Train: UnitedSynT5 for Few-Shot NLI

2024

UnitedSynT5 (3B)

First Train to Generate, then Generate to Train: UnitedSynT5 for Few-Shot NLI

2024

UnitedSynT5 (335M)

Entailment as Few-Shot Learner

2021

Neural Tree Indexers for Text Understanding

PaddlePaddle/PaddleNLP sunyilgdx/prompts4keras cactilab/hateguard

Entailment as Few-Shot Learner

2021

EFL (Entailment as Few-shot Learner) + RoBERTa-large

PaddlePaddle/PaddleNLP sunyilgdx/prompts4keras cactilab/hateguard

SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization

2019

MT-DNN-SMARTLARGEv0

namisan/mt-dnn microsoft/MT-DNN

Self-Explaining Structures Improve NLP Models

2020

RoBERTa-large+Self-Explaining

ShannonAI/Self_Explaining_Structures_Improve_NLP_Models

Self-Explaining Structures Improve NLP Models

2020

RoBERTa-large + self-explaining layer

ShannonAI/Self_Explaining_Structures_Improve_NLP_Models

Conditionally Adaptive Multi-Task Learning: Improving Transfer Learning in NLP Using Fewer Parameters & Less Data

2020

CA-MTL

CAMTL/CA-MTL

Semantics-aware BERT for Language Understanding

2019

SemBERT

cooelf/SemBERT

Multi-Task Deep Neural Networks for Natural Language Understanding

2019

MT-DNN

namisan/mt-dnn xycforgithub/MultiTask-MRC

SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization

2019

MT-DNN-SMART_100%ofTrainingData

namisan/mt-dnn microsoft/MT-DNN

Explicit Contextual Semantics for Text Comprehension

2018

SJRC (BERT-Large +SRL)

Multi-Task Deep Neural Networks for Natural Language Understanding

2019

Ntumpha

namisan/mt-dnn xycforgithub/MultiTask-MRC

Semantic Sentence Matching with Densely-connected Recurrent and Co-attentive Information

2018

Densely-Connected Recurrent and Co-Attentive Network Ensemble

Discourse Marker Augmented Network with Reinforcement Learning for Natural Language Inference

2019

300D DMAN Ensemble

ZJULearning/DMP

DR-BiLSTM: Dependent Reading Bidirectional LSTM for Natural Language Inference

2018

450D DR-BiLSTM Ensemble

Compare, Compress and Propagate: Enhancing Neural Architectures with Alignment Factorization for Natural Language Inference

2017

300D CAFE Ensemble

Deep contextualized word representations

2018

ESIM + ELMo Ensemble

flairNLP/flair dmlc/gluon-nlp

Neural Natural Language Inference Models Enhanced with External Knowledge

2017

KIM Ensemble

lukecq1231/kim MS-P3/code6

Explicit Contextual Semantics for Text Comprehension

2018

SLRC

Simple and Effective Text Matching with Richer Alignment Features

2019

RE2

alibaba-edu/simple-effective-text-matching hitvoice/RE2 alibaba-edu/simple-effective-text-matching-pytorch

Semantic Sentence Matching with Densely-connected Recurrent and Co-attentive Information

2018

Densely-Connected Recurrent and Co-Attentive Network

DEIM: An effective deep encoding and interaction model for sentence matching

2022

DEIM

Natural Language Inference over Interaction Space

2017

448D Densely Interactive Inference Network (DIIN, code) Ensemble

YichenGong/Densely-Interactive-Inference-Network YerevaNN/DIIN-in-Keras

Discourse Marker Augmented Network with Reinforcement Learning for Natural Language Inference

2019

300D DMAN

ZJULearning/DMP

Bilateral Multi-Perspective Matching for Natural Language Sentences

2017

BiMPM Ensemble

google-research-datasets/paws zhiguowang/BiMPM

Deep contextualized word representations

2018

ESIM + ELMo

flairNLP/flair dmlc/gluon-nlp

SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization

2019

MT-DNN-SMART_10%ofTrainingData

namisan/mt-dnn microsoft/MT-DNN

Neural Natural Language Inference Models Enhanced with External Knowledge

2017

KIM

lukecq1231/kim MS-P3/code6

Enhanced LSTM for Natural Language Inference

2016

600D ESIM + 300D Syntactic TreeLSTM

coetaur0/ESIM lukecq1231/nli

DR-BiLSTM: Dependent Reading Bidirectional LSTM for Natural Language Inference

2018

450D DR-BiLSTM

Stochastic Answer Networks for Natural Language Inference

2018

Stochastic Answer Network

kevinduh/san_mrc yongbowin/san_mrc_annotation Xaniar87/SAN_SQuAD2

Compare, Compress and Propagate: Enhancing Neural Architectures with Alignment Factorization for Natural Language Inference

2017

300D CAFE

Learned in Translation: Contextualized Word Vectors

2017

Biattentive Classification Network + CoVe + Char

salesforce/cove adi2103/AML-CoVe

Attention Boosted Sequential Inference Model

2018

aESIM

Natural Language Inference over Interaction Space

2017

448D Densely Interactive Inference Network (DIIN, code)

YichenGong/Densely-Interactive-Inference-Network YerevaNN/DIIN-in-Keras

Enhanced LSTM for Natural Language Inference

2016

Enhanced Sequential Inference Model (Chen et al., [2017a])

coetaur0/ESIM lukecq1231/nli

Bilateral Multi-Perspective Matching for Natural Language Sentences

2017

BiMPM

google-research-datasets/paws zhiguowang/BiMPM

Dynamic Self-Attention : Computing Attention over Words Dynamically for Sentence Embedding

2018

2400D Multiple-Dynamic Self-Attention Model

dsindex/iclassifier

Neural Tree Indexers for Text Understanding

2016

300D Full tree matching NTI-SLSTM-LSTM w/ global attention

tsendeemts/nti

Cell-aware Stacked LSTMs for Modeling Sentences

2018

300D 2-layer Bi-CAS-LSTM

A Decomposable Attention Model for Natural Language Inference

2016

200D decomposable attention feed-forward model with intra-sentence attention

dmlc/gluon-nlp harvardnlp/decomp-attn

A Decomposable Attention Model for Natural Language Inference

2016

200D decomposable attention model with intra-sentence attention

dmlc/gluon-nlp harvardnlp/decomp-attn

Dynamic Self-Attention : Computing Attention over Words Dynamically for Sentence Embedding

2018

600D Dynamic Self-Attention Model

dsindex/iclassifier

Parameter Re-Initialization through Cyclical Batch Size Schedules

2018

CBS-1 + ESIM

Dynamic Meta-Embeddings for Improved Sentence Representations

2018

512D Dynamic Meta-Embeddings

facebookresearch/DME kushalchauhan98/dynamic-meta-embeddings ellerypan/QUORA-INSINCERE-QUESTION-CLASSIFICATION

Enhancing Sentence Embedding with Generalized Pooling

2018

600D BiLSTM with generalized pooling

lukecq1231/generalized-pooling

Sentence Embeddings in NLI with Iterative Refinement Encoders

2018

600D Hierarchical BiLSTM with Max Pooling (HBMP, code)

Helsinki-NLP/HBMP

Semantic Sentence Matching with Densely-connected Recurrent and Co-attentive Information

2018

Densely-Connected Recurrent and Co-Attentive Network (encoder)

Reinforced Self-Attention Network: a Hybrid of Hard and Soft Attention for Sequence Modeling

2018

300D Reinforced Self-Attention Network

taoshen58/DiSAN

Distance-based Self-Attention Network for Natural Language Inference

2017

Distance-based Self-Attention Network

A Decomposable Attention Model for Natural Language Inference

2016

200D decomposable attention feed-forward model

dmlc/gluon-nlp harvardnlp/decomp-attn

A Decomposable Attention Model for Natural Language Inference

2016

200D decomposable attention model

dmlc/gluon-nlp harvardnlp/decomp-attn

Long Short-Term Memory-Networks for Machine Reading

2016

450D LSTMN with deep attention fusion

JRC1995/Abstractive-Summarization oneil512/lstmn muu4649/word2vec_attention

Learning Natural Language Inference with LSTM

2015

300D mLSTM word-by-word attention model

shuohangwang/SeqMatchSeq junfenglx/reasoning_attention

Learning to Compose Task-Specific Tree Structures

2017

600D Gumbel TreeLSTM encoders

jihunchoi/unsupervised-treelstm

Shortcut-Stacked Sentence Encoders for Multi-Domain Inference

2017

600D Residual stacked encoders

easonnie/multiNLI_encoder DorinK/Implementing-an-SNLI-Paper KhenAharon/Deep-Learning-SNLI-Residual-Stacked-Encoders

Star-Transformer

2019

Star-Transformer (no cross sentence attention)

dmlc/dgl fastnlp/fastNLP

SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization

2019

MT-DNN-SMART_1%ofTrainingData

namisan/mt-dnn microsoft/MT-DNN

Compare, Compress and Propagate: Enhancing Neural Architectures with Alignment Factorization for Natural Language Inference

2017

300D CAFE (no cross-sentence attention)

Shortcut-Stacked Sentence Encoders for Multi-Domain Inference

2017

300D Residual stacked encoders

easonnie/multiNLI_encoder DorinK/Implementing-an-SNLI-Paper KhenAharon/Deep-Learning-SNLI-Residual-Stacked-Encoders

Long Short-Term Memory-Networks for Machine Reading

2016

300D LSTMN with deep attention fusion

JRC1995/Abstractive-Summarization oneil512/lstmn muu4649/word2vec_attention

Learning to Compose Task-Specific Tree Structures

2017

300D Gumbel TreeLSTM encoders

jihunchoi/unsupervised-treelstm

DiSAN: Directional Self-Attention Network for RNN/CNN-Free Language Understanding

2017

300D Directional self-attention network encoders

taoshen58/DiSAN 2023-MindSpore-4/Code8 MindSpore-paper-code-3/code6

Recurrent Neural Network-Based Sentence Encoder with Gated Attention for Natural Language Inference

2017

600D (300+300) Deep Gated Attn. BiLSTM encoders

lukecq1231/enc_nli eilon47/DL_Ass4

Neural Semantic Encoders

2016

300D MMA-NSE encoders with attention

Smerity/keras_snli tsendeemts/nse saraswat/munkhdalai-nse

Modelling Interaction of Sentence Pair with coupled-LSTMs

2016

50D stacked TC-LSTMs

Learning Natural Language Inference using Bidirectional LSTM model and Inner-Attention

2016

600D (300+300) BiLSTM encoders with intra-attention and symbolic preproc.

Smerity/keras_snli songyang0716/NLP

Combining Similarity Features and Deep Representation Learning for Stance Detection in the Context of Checking Fake News

2018

Stacked Bi-LSTMs (shortcut connections, max-pooling)

imran3180/pytorch-nli LuisPB7/fnc-msc yinghao1019/imdb_prac

Neural Semantic Encoders

2016

300D NSE encoders

Smerity/keras_snli tsendeemts/nse saraswat/munkhdalai-nse

Supervised Learning of Universal Sentence Representations from Natural Language Inference Data

2017

4096D BiLSTM with max-pooling

facebookresearch/InferSent facebookresearch/SentEval

Combining Similarity Features and Deep Representation Learning for Stance Detection in the Context of Checking Fake News

2018

Bi-LSTM sentence encoder (max-pooling)

imran3180/pytorch-nli LuisPB7/fnc-msc yinghao1019/imdb_prac

Combining Similarity Features and Deep Representation Learning for Stance Detection in the Context of Checking Fake News

2018

Stacked Bi-LSTMs (shortcut connections, max-pooling, attention)

imran3180/pytorch-nli LuisPB7/fnc-msc yinghao1019/imdb_prac

Learning Natural Language Inference using Bidirectional LSTM model and Inner-Attention

2016

600D (300+300) BiLSTM encoders with intra-attention

Smerity/keras_snli songyang0716/NLP

Baseline Needs More Love: On Simple Word-Embedding-Based Models and Associated Pooling Mechanisms

2018

SWEM-max

dinghanshen/SWEM nyk510/scdv-python

Reasoning about Entailment with Neural Attention

2015

100D LSTMs w/ word-by-word attention

shyamupa/snli-entailment junfenglx/reasoning_attention

Neural Tree Indexers for Text Understanding

2016

300D NTI-SLSTM-LSTM encoders

tsendeemts/nti

Learning Natural Language Inference using Bidirectional LSTM model and Inner-Attention

2016

600D (300+300) BiLSTM encoders

Smerity/keras_snli songyang0716/NLP

A Fast Unified Model for Parsing and Sentence Understanding

2016

300D SPINN-PI encoders

stanfordnlp/spinn NYU-MLL/spinn richa912/THANOS

SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization

2019

MT-DNN-SMART_0.1%ofTrainingData

namisan/mt-dnn microsoft/MT-DNN

Natural Language Inference by Tree-Based Convolution and Heuristic Matching

2015

300D Tree-based CNN encoders

Order-Embeddings of Images and Language

2015

1024D GRU encoders w/ unsupervised 'skip-thoughts' pre-training

ivendrov/order-embedding iesl/geometric_graph_embedding

DELTA: A DEep learning based Language Technology plAtform

2019

DELTA (LSTM)

didi/delta Delta-ML/delta

A Fast Unified Model for Parsing and Sentence Understanding

2016

300D LSTM encoders

stanfordnlp/spinn NYU-MLL/spinn richa912/THANOS

SplitEE: Early Exit in Deep Neural Networks with Split Computing

2023

SplitEE-S

Div290/SplitEE

A large annotated corpus for learning natural language inference

2015

+ Unigram and bigram features

kawine/dataset_difficulty hpprc/simple-simcse-ja songyang0716/NLP

A large annotated corpus for learning natural language inference

2015

100D LSTM encoders

kawine/dataset_difficulty hpprc/simple-simcse-ja songyang0716/NLP

A large annotated corpus for learning natural language inference

2015

Unlexicalized features

kawine/dataset_difficulty hpprc/simple-simcse-ja songyang0716/NLP

Model	Paper	% Test Accuracy	Date
UnitedSynT5 (3B)	First Train to Generate, then Generate to Train: …	94.70	2024-12-12
UnitedSynT5 (335M)	First Train to Generate, then Generate to Train: …	93.50	2024-12-12
Neural Tree Indexers for Text Understanding	Entailment as Few-Shot Learner	93.10	2021-04-29
EFL (Entailment as Few-shot Learner) + RoBERTa-large	Entailment as Few-Shot Learner	93.10	2021-04-29
MT-DNN-SMARTLARGEv0	SMART: Robust and Efficient Fine-Tuning for Pre-t…	92.60	2019-11-08
RoBERTa-large+Self-Explaining	Self-Explaining Structures Improve NLP Models	92.30	2020-12-03
RoBERTa-large + self-explaining layer	Self-Explaining Structures Improve NLP Models	92.30	2020-12-03
CA-MTL	Conditionally Adaptive Multi-Task Learning: Impro…	92.10	2020-09-19
SemBERT	Semantics-aware BERT for Language Understanding	91.90	2019-09-05
MT-DNN	Multi-Task Deep Neural Networks for Natural Langu…	91.60	2019-01-31
MT-DNN-SMART_100%ofTrainingData	SMART: Robust and Efficient Fine-Tuning for Pre-t…	91.60	2019-11-08
SJRC (BERT-Large +SRL)	Explicit Contextual Semantics for Text Comprehens…	91.30	2018-09-08
Ntumpha	Multi-Task Deep Neural Networks for Natural Langu…	90.50	2019-01-31
Densely-Connected Recurrent and Co-Attentive Network Ensemble	Semantic Sentence Matching with Densely-connected…	90.10	2018-05-29
300D DMAN Ensemble	Discourse Marker Augmented Network with Reinforce…	89.60	2019-07-23
450D DR-BiLSTM Ensemble	DR-BiLSTM: Dependent Reading Bidirectional LSTM f…	89.30	2018-02-15
300D CAFE Ensemble	Compare, Compress and Propagate: Enhancing Neural…	89.30	2017-12-30
ESIM + ELMo Ensemble	Deep contextualized word representations	89.30	2018-02-15
KIM Ensemble	Neural Natural Language Inference Models Enhanced…	89.10	2017-11-12
SLRC	Explicit Contextual Semantics for Text Comprehens…	89.10	2018-09-08
RE2	Simple and Effective Text Matching with Richer Al…	88.90	2019-08-01
Densely-Connected Recurrent and Co-Attentive Network	Semantic Sentence Matching with Densely-connected…	88.90	2018-05-29
DEIM	DEIM: An effective deep encoding and interaction …	88.90	2022-03-20
448D Densely Interactive Inference Network (DIIN, code) Ensemble	Natural Language Inference over Interaction Space	88.90	2017-09-13
300D DMAN	Discourse Marker Augmented Network with Reinforce…	88.80	2019-07-23
BiMPM Ensemble	Bilateral Multi-Perspective Matching for Natural …	88.80	2017-02-13
ESIM + ELMo	Deep contextualized word representations	88.70	2018-02-15
MT-DNN-SMART_10%ofTrainingData	SMART: Robust and Efficient Fine-Tuning for Pre-t…	88.70	2019-11-08
KIM	Neural Natural Language Inference Models Enhanced…	88.60	2017-11-12
600D ESIM + 300D Syntactic TreeLSTM	Enhanced LSTM for Natural Language Inference	88.60	2016-09-20
450D DR-BiLSTM	DR-BiLSTM: Dependent Reading Bidirectional LSTM f…	88.50	2018-02-15
Stochastic Answer Network	Stochastic Answer Networks for Natural Language I…	88.50	2018-04-21
300D CAFE	Compare, Compress and Propagate: Enhancing Neural…	88.50	2017-12-30
Biattentive Classification Network + CoVe + Char	Learned in Translation: Contextualized Word Vecto…	88.10	2017-08-01
aESIM	Attention Boosted Sequential Inference Model	88.10	2018-12-05
448D Densely Interactive Inference Network (DIIN, code)	Natural Language Inference over Interaction Space	88.00	2017-09-13
Enhanced Sequential Inference Model (Chen et al., [2017a])	Enhanced LSTM for Natural Language Inference	88.00	2016-09-20
BiMPM	Bilateral Multi-Perspective Matching for Natural …	87.50	2017-02-13
2400D Multiple-Dynamic Self-Attention Model	Dynamic Self-Attention : Computing Attention over…	87.40	2018-08-22
300D Full tree matching NTI-SLSTM-LSTM w/ global attention	Neural Tree Indexers for Text Understanding	87.30	2016-07-15
300D 2-layer Bi-CAS-LSTM	Cell-aware Stacked LSTMs for Modeling Sentences	87.00	2018-09-07
200D decomposable attention feed-forward model with intra-sentence attention	A Decomposable Attention Model for Natural Langua…	86.80	2016-06-06
200D decomposable attention model with intra-sentence attention	A Decomposable Attention Model for Natural Langua…	86.80	2016-06-06
600D Dynamic Self-Attention Model	Dynamic Self-Attention : Computing Attention over…	86.80	2018-08-22
CBS-1 + ESIM	Parameter Re-Initialization through Cyclical Batc…	86.73	2018-12-04
512D Dynamic Meta-Embeddings	Dynamic Meta-Embeddings for Improved Sentence Rep…	86.70	2018-04-21
600D BiLSTM with generalized pooling	Enhancing Sentence Embedding with Generalized Poo…	86.60	2018-06-26
600D Hierarchical BiLSTM with Max Pooling (HBMP, code)	Sentence Embeddings in NLI with Iterative Refinem…	86.60	2018-08-27
Densely-Connected Recurrent and Co-Attentive Network (encoder)	Semantic Sentence Matching with Densely-connected…	86.50	2018-05-29
300D Reinforced Self-Attention Network	Reinforced Self-Attention Network: a Hybrid of Ha…	86.30	2018-01-31
Distance-based Self-Attention Network	Distance-based Self-Attention Network for Natural…	86.30	2017-12-06
200D decomposable attention feed-forward model	A Decomposable Attention Model for Natural Langua…	86.30	2016-06-06
200D decomposable attention model	A Decomposable Attention Model for Natural Langua…	86.30	2016-06-06
450D LSTMN with deep attention fusion	Long Short-Term Memory-Networks for Machine Readi…	86.30	2016-01-25
300D mLSTM word-by-word attention model	Learning Natural Language Inference with LSTM	86.10	2015-12-30
600D Gumbel TreeLSTM encoders	Learning to Compose Task-Specific Tree Structures	86.00	2017-07-10
600D Residual stacked encoders	Shortcut-Stacked Sentence Encoders for Multi-Doma…	86.00	2017-08-07
Star-Transformer (no cross sentence attention)	Star-Transformer	86.00	2019-02-25
MT-DNN-SMART_1%ofTrainingData	SMART: Robust and Efficient Fine-Tuning for Pre-t…	86.00	2019-11-08
300D CAFE (no cross-sentence attention)	Compare, Compress and Propagate: Enhancing Neural…	85.90	2017-12-30
300D Residual stacked encoders	Shortcut-Stacked Sentence Encoders for Multi-Doma…	85.70	2017-08-07
300D LSTMN with deep attention fusion	Long Short-Term Memory-Networks for Machine Readi…	85.70	2016-01-25
300D Gumbel TreeLSTM encoders	Learning to Compose Task-Specific Tree Structures	85.60	2017-07-10
300D Directional self-attention network encoders	DiSAN: Directional Self-Attention Network for RNN…	85.60	2017-09-14
600D (300+300) Deep Gated Attn. BiLSTM encoders	Recurrent Neural Network-Based Sentence Encoder w…	85.50	2017-08-04
300D MMA-NSE encoders with attention	Neural Semantic Encoders	85.40	2016-07-14
50D stacked TC-LSTMs	Modelling Interaction of Sentence Pair with coupl…	85.10	2016-05-18
600D (300+300) BiLSTM encoders with intra-attention and symbolic preproc.	Learning Natural Language Inference using Bidirec…	85.00	2016-05-30
Stacked Bi-LSTMs (shortcut connections, max-pooling)	Combining Similarity Features and Deep Representa…	84.80	2018-11-02
300D NSE encoders	Neural Semantic Encoders	84.60	2016-07-14
4096D BiLSTM with max-pooling	Supervised Learning of Universal Sentence Represe…	84.50	2017-05-05
Bi-LSTM sentence encoder (max-pooling)	Combining Similarity Features and Deep Representa…	84.50	2018-11-02
Stacked Bi-LSTMs (shortcut connections, max-pooling, attention)	Combining Similarity Features and Deep Representa…	84.40	2018-11-02
600D (300+300) BiLSTM encoders with intra-attention	Learning Natural Language Inference using Bidirec…	84.20	2016-05-30
SWEM-max	Baseline Needs More Love: On Simple Word-Embeddin…	83.80	2018-05-24
100D LSTMs w/ word-by-word attention	Reasoning about Entailment with Neural Attention	83.50	2015-09-22
300D NTI-SLSTM-LSTM encoders	Neural Tree Indexers for Text Understanding	83.40	2016-07-15
600D (300+300) BiLSTM encoders	Learning Natural Language Inference using Bidirec…	83.30	2016-05-30
300D SPINN-PI encoders	A Fast Unified Model for Parsing and Sentence Und…	83.20	2016-03-19
MT-DNN-SMART_0.1%ofTrainingData	SMART: Robust and Efficient Fine-Tuning for Pre-t…	82.70	2019-11-08
300D Tree-based CNN encoders	Natural Language Inference by Tree-Based Convolut…	82.10	2015-12-28
1024D GRU encoders w/ unsupervised 'skip-thoughts' pre-training	Order-Embeddings of Images and Language	81.40	2015-11-19
DELTA (LSTM)	DELTA: A DEep learning based Language Technology …	80.70	2019-08-02
300D LSTM encoders	A Fast Unified Model for Parsing and Sentence Und…	80.60	2016-03-19
SplitEE-S	SplitEE: Early Exit in Deep Neural Networks with …	79.00	2023-09-17
+ Unigram and bigram features	A large annotated corpus for learning natural lan…	78.20	2015-08-21
100D LSTM encoders	A large annotated corpus for learning natural lan…	77.60	2015-08-21
Unlexicalized features	A large annotated corpus for learning natural lan…	50.40	2015-08-21

SNLI

Performance Over Time

Edit Benchmark Results

Edit Result

Top Performing Models

All Papers (88)