Natural Language Inference on SNLI

Metric: % Test Accuracy (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

#	Model↕	% Test Accuracy▼	Extra Data	Paper	Date↕	Code
1	UnitedSynT5 (3B)	94.7	Yes	First Train to Generate, then Generate to Train:...	2024-12-12	-
2	UnitedSynT5 (335M)	93.5	Yes	First Train to Generate, then Generate to Train:...	2024-12-12	-
3	Neural Tree Indexers for Text Understanding	93.1	No	Entailment as Few-Shot Learner	2021-04-29	Code
4	EFL (Entailment as Few-shot Learner) + RoBERTa-large	93.1	No	Entailment as Few-Shot Learner	2021-04-29	Code
5	RoBERTa-large+Self-Explaining	92.3	No	Self-Explaining Structures Improve NLP Models	2020-12-03	Code
6	RoBERTa-large + self-explaining layer	92.3	No	Self-Explaining Structures Improve NLP Models	2020-12-03	Code
7	CA-MTL	92.1	No	Conditionally Adaptive Multi-Task Learning: Impr...	2020-09-19	Code
8	SemBERT	91.9	No	Semantics-aware BERT for Language Understanding	2019-09-05	Code
9	MT-DNN-SMARTLARGEv0	91.7	No	SMART: Robust and Efficient Fine-Tuning for Pre-...	2019-11-08	Code
10	MT-DNN	91.6	No	Multi-Task Deep Neural Networks for Natural Lang...	2019-01-31	Code
11	SJRC (BERT-Large +SRL)	91.3	No	Explicit Contextual Semantics for Text Comprehen...	2018-09-08	-
12	Ntumpha	90.5	No	Multi-Task Deep Neural Networks for Natural Lang...	2019-01-31	Code
13	Densely-Connected Recurrent and Co-Attentive Network Ensemble	90.1	No	Semantic Sentence Matching with Densely-connecte...	2018-05-29	-
14	MFAE	90.07	No	-	-	Code
15	Fine-Tuned LM-Pretrained Transformer	89.9	No	-	-	Code
16	300D DMAN Ensemble	89.6	No	Discourse Marker Augmented Network with Reinforc...	2019-07-23	Code
17	300D DMAN Ensemble	89.6	No	Discourse Marker Augmented Network with Reinforc...	2019-07-23	Code
18	150D Multiway Attention Network Ensemble	89.4	No	-	-	Code
19	450D DR-BiLSTM Ensemble	89.3	No	DR-BiLSTM: Dependent Reading Bidirectional LSTM ...	2018-02-15	-
20	300D CAFE Ensemble	89.3	No	Compare, Compress and Propagate: Enhancing Neura...	2017-12-30	-
21	ESIM + ELMo Ensemble	89.3	No	Deep contextualized word representations	2018-02-15	Code
22	KIM Ensemble	89.1	No	Neural Natural Language Inference Models Enhance...	2017-11-12	Code
23	SLRC	89.1	No	Explicit Contextual Semantics for Text Comprehen...	2018-09-08	-
24	RE2	88.9	No	Simple and Effective Text Matching with Richer A...	2019-08-01	Code
25	Densely-Connected Recurrent and Co-Attentive Network	88.9	No	Semantic Sentence Matching with Densely-connecte...	2018-05-29	-
26	DEIM	88.9	No	DEIM: An effective deep encoding and interaction...	2022-03-20	-
27	448D Densely Interactive Inference Network (DIIN, code) Ensemble	88.9	No	Natural Language Inference over Interaction Space	2017-09-13	Code
28	300D DMAN	88.8	No	Discourse Marker Augmented Network with Reinforc...	2019-07-23	Code
29	300D DMAN	88.8	No	Discourse Marker Augmented Network with Reinforc...	2019-07-23	Code
30	BiMPM Ensemble	88.8	No	Bilateral Multi-Perspective Matching for Natural...	2017-02-13	Code
31	ESIM + ELMo	88.7	No	Deep contextualized word representations	2018-02-15	Code
32	KIM	88.6	No	Neural Natural Language Inference Models Enhance...	2017-11-12	Code
33	600D ESIM + 300D Syntactic TreeLSTM	88.6	No	Enhanced LSTM for Natural Language Inference	2016-09-20	Code
34	450D DR-BiLSTM	88.5	No	DR-BiLSTM: Dependent Reading Bidirectional LSTM ...	2018-02-15	-
35	Stochastic Answer Network	88.5	No	Stochastic Answer Networks for Natural Language ...	2018-04-21	Code
36	300D CAFE	88.5	No	Compare, Compress and Propagate: Enhancing Neura...	2017-12-30	-
37	150D Multiway Attention Network	88.3	No	-	-	Code
38	Biattentive Classification Network + CoVe + Char	88.1	No	Learned in Translation: Contextualized Word Vect...	2017-08-01	Code
39	aESIM	88.1	No	Attention Boosted Sequential Inference Model	2018-12-05	-
40	448D Densely Interactive Inference Network (DIIN, code)	88	No	Natural Language Inference over Interaction Space	2017-09-13	Code
41	Enhanced Sequential Inference Model (Chen et al., [2017a])	88	No	Enhanced LSTM for Natural Language Inference	2016-09-20	Code
42	BiMPM	87.5	No	Bilateral Multi-Perspective Matching for Natural...	2017-02-13	Code
43	300D re-read LSTM	87.5	No	-	-	-
44	300D re-read LSTM	87.5	No	-	-	-
45	2400D Multiple-Dynamic Self-Attention Model	87.4	No	Dynamic Self-Attention : Computing Attention ove...	2018-08-22	Code
46	300D Full tree matching NTI-SLSTM-LSTM w/ global attention	87.3	No	Neural Tree Indexers for Text Understanding	2016-07-15	Code
47	300D 2-layer Bi-CAS-LSTM	87	No	Cell-aware Stacked LSTMs for Modeling Sentences	2018-09-07	-
48	200D decomposable attention feed-forward model with intra-sentence attention	86.8	No	A Decomposable Attention Model for Natural Langu...	2016-06-06	Code
49	200D decomposable attention model with intra-sentence attention	86.8	No	A Decomposable Attention Model for Natural Langu...	2016-06-06	Code
50	600D Dynamic Self-Attention Model	86.8	No	Dynamic Self-Attention : Computing Attention ove...	2018-08-22	Code
51	CBS-1 + ESIM	86.73	No	Parameter Re-Initialization through Cyclical Bat...	2018-12-04	-
52	512D Dynamic Meta-Embeddings	86.7	No	Dynamic Meta-Embeddings for Improved Sentence Re...	2018-04-21	Code
53	600D BiLSTM with generalized pooling	86.6	No	Enhancing Sentence Embedding with Generalized Po...	2018-06-26	Code
54	600D Hierarchical BiLSTM with Max Pooling (HBMP, code)	86.6	No	Sentence Embeddings in NLI with Iterative Refine...	2018-08-27	Code
55	Densely-Connected Recurrent and Co-Attentive Network (encoder)	86.5	No	Semantic Sentence Matching with Densely-connecte...	2018-05-29	-
56	300D Reinforced Self-Attention Network	86.3	No	Reinforced Self-Attention Network: a Hybrid of H...	2018-01-31	Code
57	Distance-based Self-Attention Network	86.3	No	Distance-based Self-Attention Network for Natura...	2017-12-06	-
58	200D decomposable attention feed-forward model	86.3	No	A Decomposable Attention Model for Natural Langu...	2016-06-06	Code
59	200D decomposable attention model	86.3	No	A Decomposable Attention Model for Natural Langu...	2016-06-06	Code
60	450D LSTMN with deep attention fusion	86.3	No	Long Short-Term Memory-Networks for Machine Read...	2016-01-25	Code
61	300D mLSTM word-by-word attention model	86.1	No	Learning Natural Language Inference with LSTM	2015-12-30	Code
62	600D Gumbel TreeLSTM encoders	86	No	Learning to Compose Task-Specific Tree Structures	2017-07-10	Code
63	600D Residual stacked encoders	86	No	Shortcut-Stacked Sentence Encoders for Multi-Dom...	2017-08-07	Code
64	Star-Transformer (no cross sentence attention)	86	No	Star-Transformer	2019-02-25	Code
65	300D CAFE (no cross-sentence attention)	85.9	No	Compare, Compress and Propagate: Enhancing Neura...	2017-12-30	-
66	1200D REGMAPR (Base+Reg)	85.9	No	-	-	-
67	300D Residual stacked encoders	85.7	No	Shortcut-Stacked Sentence Encoders for Multi-Dom...	2017-08-07	Code
68	300D LSTMN with deep attention fusion	85.7	No	Long Short-Term Memory-Networks for Machine Read...	2016-01-25	Code
69	300D Gumbel TreeLSTM encoders	85.6	No	Learning to Compose Task-Specific Tree Structures	2017-07-10	Code
70	300D Directional self-attention network encoders	85.6	No	DiSAN: Directional Self-Attention Network for RN...	2017-09-14	Code
71	600D (300+300) Deep Gated Attn. BiLSTM encoders	85.5	No	Recurrent Neural Network-Based Sentence Encoder ...	2017-08-04	Code
72	300D MMA-NSE encoders with attention	85.4	No	Neural Semantic Encoders	2016-07-14	Code
73	50D stacked TC-LSTMs	85.1	No	Modelling Interaction of Sentence Pair with coup...	2016-05-18	-
74	600D (300+300) BiLSTM encoders with intra-attention and symbolic preproc.	85	No	Learning Natural Language Inference using Bidire...	2016-05-30	Code
75	Stacked Bi-LSTMs (shortcut connections, max-pooling)	84.8	No	Combining Similarity Features and Deep Represent...	2018-11-02	Code
76	300D NSE encoders	84.6	No	Neural Semantic Encoders	2016-07-14	Code
77	100D DF-LSTM	84.6	No	-	-	-
78	4096D BiLSTM with max-pooling	84.5	No	Supervised Learning of Universal Sentence Repres...	2017-05-05	Code
79	Bi-LSTM sentence encoder (max-pooling)	84.5	No	Combining Similarity Features and Deep Represent...	2018-11-02	Code
80	Stacked Bi-LSTMs (shortcut connections, max-pooling, attention)	84.4	No	Combining Similarity Features and Deep Represent...	2018-11-02	Code
81	600D (300+300) BiLSTM encoders with intra-attention	84.2	No	Learning Natural Language Inference using Bidire...	2016-05-30	Code
82	SWEM-max	83.8	No	Baseline Needs More Love: On Simple Word-Embeddi...	2018-05-24	Code
83	100D LSTMs w/ word-by-word attention	83.5	No	Reasoning about Entailment with Neural Attention	2015-09-22	Code
84	300D NTI-SLSTM-LSTM encoders	83.4	No	Neural Tree Indexers for Text Understanding	2016-07-15	Code
85	600D (300+300) BiLSTM encoders	83.3	No	Learning Natural Language Inference using Bidire...	2016-05-30	Code
86	300D SPINN-PI encoders	83.2	No	A Fast Unified Model for Parsing and Sentence Un...	2016-03-19	Code
87	300D Tree-based CNN encoders	82.1	No	Natural Language Inference by Tree-Based Convolu...	2015-12-28	-
88	1024D GRU encoders w/ unsupervised 'skip-thoughts' pre-training	81.4	No	Order-Embeddings of Images and Language	2015-11-19	Code
89	DELTA (LSTM)	80.7	No	DELTA: A DEep learning based Language Technology...	2019-08-02	Code
90	300D LSTM encoders	80.6	No	A Fast Unified Model for Parsing and Sentence Un...	2016-03-19	Code
91	+ Unigram and bigram features	78.2	No	A large annotated corpus for learning natural la...	2015-08-21	Code
92	100D LSTM encoders	77.6	No	A large annotated corpus for learning natural la...	2015-08-21	Code
93	Unlexicalized features	50.4	No	A large annotated corpus for learning natural la...	2015-08-21	Code