Natural Language Inference on XNLI Hi (dev)
76.91AccuracySkyline
Evaluation Results
| Method | Links | |
|---|---|---|
| SkylineBackbone=XLMR-large, Input Language=en2026.04 | 76.91 | |
| T-top-kEvaluation protocol=translate-train, Supervision type=teacher soft pseudolabels, Data selection strategy=top-k2024.07 | 71.55 | |
| Tpl-rand-kselection strategy=rand-k, Zero-shot=true2024.07 | 71.49 | |
| Tpl-div-kEvaluation protocol=translate-train, Supervision type=hard prompt labels, Data selection strategy=div-k2024.07 | 71.31 | |
| T-div-kEvaluation protocol=translate-train, Supervision type=teacher soft pseudolabels, Data selection strategy=div-k2024.07 | 71.26 | |
| T-rand-kEvaluation protocol=translate-train, Supervision type=teacher soft pseudolabels, Data selection strategy=rand-k2024.07 | 71.05 | |
| Tpl-rand-kEvaluation protocol=translate-train, Supervision type=hard prompt labels, Data selection strategy=rand-k2024.07 | 71.05 | |
| T-top-kselection strategy=top-k, Zero-shot=true2024.07 | 70.82 | |
| Tpl-top-kEvaluation protocol=translate-train, Supervision type=hard prompt labels, Data selection strategy=top-k2024.07 | 70.4 | |
| T-div-kselection strategy=div-k, Zero-shot=true2024.07 | 70.38 | |
| SRC+GENZero-shot=true2024.07 | 70.1 | |
| Tpl-top-kselection strategy=top-k, Zero-shot=true2024.07 | 70.02 | |
| Tpl-div-kselection strategy=div-k, Zero-shot=true2024.07 | 70 | |
| SRC+GENEvaluation protocol=translate-train2024.07 | 69.74 | |
| T-rand-kselection strategy=rand-k, Zero-shot=true2024.07 | 68.8 | |
| SRCEvaluation protocol=translate-train2024.07 | 67.67 | |
| SkylineBackbone=XLMR-large, Input Language=hi2026.04 | 66.12 | |
| SRCZero-shot=true2024.07 | 65.54 | |
| GENZero-shot=true2024.07 | 62.89 | |
| GENEvaluation protocol=translate-train2024.07 | 59.98 | |
| XITE-reg-ldaBackbone=XLMR-large, Input Language=en2026.04 | 57.35 | |
| XITE-reg-regBackbone=XLMR-large, Input Language=en2026.04 | 55.46 | |
| XITE-reg-ldaBackbone=XLMR-large, Input Language=hi2026.04 | 51.02 | |
| XITE-reg-regBackbone=XLMR-large, Input Language=hi2026.04 | 42.45 | |
| Baseline-PSBackbone=XLMR-large, Input Language=hi2026.04 | 35.31 | |
| Baseline-PSBackbone=XLMR-large, Input Language=en2026.04 | 33.78 |