Natural Language Understanding on GLUE small benchmark leaderboard · SOTA2 Research

Natural Language Understanding on GLUE small

73.4CoLA MccARMADA

Evaluation Results

Method	Links
ARMADA2026.03		73.4	89.4	91.6	92
ARMADA2026.03		73.1	90.2	91.6	92.3
ARMADA2026.03		72.6	89.9	92.3	92.4
DeBERTa-v2-xxlarge2026.03		72.1	89.7	90.3	92.3
undistilled2026.03		72.1	89.7	90.3	92.3
ARMADA2026.03		71.4	90.3	90.9	91.9
ARMADA2026.03		70.4	90.5	90.3	92.1
ARMADA2026.03		68.2	91.1	90.8	91.8
ARMADA2026.03		62.6	84.5	77.3	90.4
ARMADA2026.03		62.4	84.7	78.6	90
ARMADA2026.03		61.9	85.2	73.9	89.7
ARMADA2026.03		61.8	85	71.7	89.6
ARMADA2026.03		61.7	83.1	79	90.1
ARMADA2026.03		61.6	85	72.5	89.5
ARMADA2026.03		61.2	82.4	77.1	89.9
ARMADA2026.03		61.1	85	68.8	88.3
BERT-base2026.03		60.8	83.8	71	89.3
undistilled2026.03		60.8	83.8	71	89.3
ARMADA2026.03		60.4	83.4	74.2	90
ARMADA2026.03		60.1	84.1	68.8	88.3
OPT-1.3B2026.03		59.6	84.3	76.3	89.7
undistilled2026.03		59.6	84.3	76.3	89.7
ARMADA2026.03		59.5	80.5	77.1	90.1
ARMADA2026.03		58.8	84.9	69.9	88.6
ARMADA2026.03		47.6	80.5	67.4	88.5
ARMADA2026.03		46.9	79.3	67	88.3
ARMADA2026.03		46.1	80.3	67.5	88.5
ARMADA2026.03		45.7	79.5	67.5	88.6
ARMADA2026.03		44	80.1	65.2	87.9
ARMADA2026.03		42.9	79.1	66.3	88.3
BERT-6L2026.03		42.8	78.6	64.1	88.5
undistilled2026.03		42.8	78.6	64.1	88.5