Natural Language Understanding on MRPC, QQP, STS-B, MNLI, QNLI, RTE, SNLI, Sci, SICK, and Twi benchmark leaderboard · SOTA2 Research

Natural Language Understanding on MRPC, QQP, STS-B, MNLI, QNLI, RTE, SNLI, Sci, SICK, and Twi

90.2MRPC AccuracyRoBERTa-Large-MGRN

Evaluation Results

Method	Links
RoBERTa-Large-MGRN2026.04		90.2	91.2	90.6	90.2	90.1	94.2	84.1	91.2	94.2	91.4	92.3	90.7
BERT-Large-MGRN2026.04		89.9	90.2	88.1	86.8	86.7	93	72	91.1	94.3	91.2	92.4	88.8
RoBERTa-Base-MGRN2026.04		89.5	91.2	88.1	87.7	87.5	93.2	82.5	91.2	93.1	89.5	87.6	89.2
RoBERTa-Large2026.04		89.4	89.7	90.2	89.5	89.3	92.7	83.8	91.2	94.3	91.2	91.9	90.3
BERT-Base-MGRN2026.04		89.3	89.1	87.1	85.1	84.9	91.2	68.5	91.2	92.1	87.8	86.6	86.8
RoBERTa-Base2026.04		89.3	89.6	87.4	86.3	86.2	92.2	73.6	90.8	92.3	87.9	85.9	87.6
SyntaxBERT2026.04		89.2	89.6	88.1	84.9	84.6	91.1	68.9	91	92.7	88.7	87.3	86.3
MGRN2026.04		89.1	91.3	88.2	84.9	84.7	91.4	69.5	91.3	93.6	88.6	87.5	86.7
BERT-Large2026.04		88.9	89.3	87.6	86.8	86.3	92.7	70.1	91	94.4	91.1	91.5	88
UERBERT2026.04		88.3	90.5	85.1	84.2	83.5	90.6	67.1	90.8	92.2	87.8	86.2	86
SemBERT2026.04		88.2	90.2	87.3	84.4	84	90.9	69.3	90.9	92.5	87.9	86.8	86.5
BERT-Base2026.04		87.2	89.1	86.8	84.3	83.7	90.4	67.2	90.7	91.8	87.2	84.8	85.8
BiLSTM+ELMo+Attnt2026.04		84.6	86.7	73.3	76.4	76.1	79.8	56.8	89	85.8	78.9	81.4	78.9
CAFE2026.04		82.4	88	—	78.7	77.9	81.5	56.8	88.5	83.3	72.3	—	—
OpenAI GPT2026.04		82.3	81.3	80	82.1	81.4	87.4	56	88.4	84.8	79.5	81.9	80.4
Transformer2026.04		81.7	84.4	73.6	72.3	71.4	80.3	58	84.6	72.9	70.3	68.8	74.4
ESIM2026.04		80.3	88.2	—	75.8	75.6	80.5	—	88	70.6	71.8	—	—
BiMPM2026.04		79.6	85	—	72.3	72.1	81.4	56.4	—	—	—	—	—