OpenCodePapers

natural-language-inference-on-qnli-glue-dev

Natural Language Inference
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Click a sortable column header to sort.
PaperCodeAccuracyModelNameReleaseDate
DeBERTa: Decoding-enhanced BERT with Disentangled Attention✓ Link95.3DeBERTa Large2020-06-05
MVP: Multi-task Supervised Pre-training for Natural Language Generation✓ Link95.10MVP2022-06-24
ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators✓ Link95.0ELECTRA-1.75M2020-03-23
Larger-Scale Transformers for Multilingual Masked Language Modeling✓ Link95.0XLM-R XXL2021-05-02
MPNet: Masked and Permuted Pre-training for Language Understanding✓ Link94.9MPNet Large2020-04-20
ReFTA: Breaking the Weight Reconstruction Bottleneck in Tensorized Parameter-Efficient Fine-Tuning✓ Link94.8ReFTA (RoBERTa-Large)2026-06-03
RoBERTa: A Robustly Optimized BERT Pretraining Approach✓ Link94.7RoBERTa2019-07-26
I-BERT: Integer-only BERT Quantization✓ Link94.5I-BERT RoBERTa-Large2021-01-05
LFM2.5-Encoders: Fast at Long Context, Even on CPU✓ Link93.97LFM2.5-Encoder-350M2026-07-28
Unsupervised Cross-lingual Representation Learning at Scale✓ Link93.8XLM-R2019-11-05
ERNIE-Gram: Pre-Training with Explicitly N-Gram Masked Language Modeling for Natural Language Understanding✓ Link93.2ERNIE-Gram2020-10-23
GLU Variants Improve Transformer✓ Link92.93FFN SwiGLU2020-02-12
Resource-Efficient Transformer Pruning for Finetuning of Large Models✓ Link90.74RECAP (BERT-base, 50% pruning)2024-06-17
You Only Sample (Almost) Once: Linear Cost Self-Attention Via Bernoulli Sampling✓ Link90.1YOSO-E2021-11-18
DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter✓ Link89.2DistilBERT2019-10-02
RoFormer: Enhanced Transformer with Rotary Position Embedding✓ Link88.0RoFormer2021-04-20
FLAVA: A Foundational Language And Vision Alignment Model✓ Link87.3FLAVA2021-12-08
Check out how to contribute new results. Then edit this file on GitLab or GitHub.