Aspect-based Sentiment Quadruple Extraction on LASQ Uzbek
75.9Entity Target F1SKEM
Evaluation results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| SKEM2026.04 | 75.9 | 28.44 | 54.86 | 22.27 | 40.79 | 20.45 | 22.71 | 13.82 | 17.18 | |
| H2DT2026.04 | 75.77 | 25.17 | 54.65 | 19.69 | 40.36 | 19.96 | 24.3 | 11.76 | 15.85 | |
| Meta-WP2026.04 | 75.62 | 25.35 | 54.51 | 20.57 | 40.4 | 19.29 | 25.52 | 11.13 | 15.5 | |
| MRM-STS2026.04 | 73.52 | 26.52 | 50.57 | 21.05 | 39.7 | 20.26 | 18.66 | 12.82 | 15.9 | |
| Span-ASTE2026.04 | 65.21 | 24.29 | 46.81 | 17.92 | 32.17 | 16.83 | 18.96 | 10.31 | 13.36 | |
| EC-ACOS2026.04 | 64.06 | 23.32 | 44.75 | 16.41 | 30.92 | 15.16 | 16.94 | 8.72 | 11.51 | |
| ParaPhrase2026.04 | 63.16 | 22.86 | 47.87 | 16.91 | 33.5 | 15.89 | 22.27 | 8.96 | 12.78 | |
| GPT-5Evaluation Setting=In-context learning (ICL)2026.04 | 40.24 | 16.56 | 30.53 | 15.35 | 27.02 | 16.3 | 20.23 | 9.5 | 12.93 | |
| Qwen3-MaxEvaluation Setting=In-context learning (ICL)2026.04 | 40.06 | 16.86 | 30.25 | 14.29 | 30.21 | 15.38 | 18.64 | 8.04 | 11.23 | |
| GPT-5Evaluation Setting=Zero-shot (ZS)2026.04 | 38.32 | 13.89 | 27.75 | 13.71 | 25.11 | 14.55 | 19.3 | 6.52 | 9.75 | |
| Qwen3-MaxEvaluation Setting=Zero-shot (ZS)2026.04 | 35.51 | 14.05 | 26.22 | 11.01 | 26.01 | 12.31 | 15.53 | 5.91 | 8.56 |