Coreference Resolution on CIN (test)
49.67MUC RecallClaude3
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Claude3Params=Unk.2026.06 | 49.67 | 39.66 | 41.21 | 90.04 | 85.5 | 86.58 | 76.86 | 88.17 | 80.91 | 69.57 | |
| PA-MCRParams=149M2026.06 | 41.79 | 42.42 | 39.28 | 90.31 | 91.37 | 90.31 | 85.91 | 88.11 | 86.48 | 72.02 | |
| SA-MCRParams=209M2026.06 | 39.83 | 44.7 | 41.43 | 72.42 | 90.27 | 80.32 | 66.51 | 96.63 | 78.38 | 66.71 | |
| Qwen2.5Params=8B2026.06 | 39.63 | 39.49 | 36.67 | 82.72 | 85.76 | 83.66 | 70.56 | 87.82 | 77.46 | 65.93 | |
| CLIPParams=149M2026.06 | 38.08 | 16.99 | 21.67 | 89.71 | 61.98 | 70.99 | 51.37 | 75.2 | 59.27 | 50.64 | |
| QWen2.5-VLParams=8B2026.06 | 37.48 | 40.71 | 36.32 | 86.65 | 92.41 | 89 | 83.11 | 87.16 | 84.38 | 69.9 | |
| VR-CLIPParams=149M2026.06 | 33.28 | 22.72 | 25.2 | 88.28 | 77.19 | 81.6 | 69.02 | 82.04 | 74.29 | 60.36 | |
| Semi-MCRParams=209M2026.06 | 31.11 | 35.25 | 31.86 | 70.63 | 87.85 | 78.06 | 63.99 | 93.44 | 75.47 | 61.79 | |
| WS-MCRParams=44M2026.06 | 24.87 | 18.34 | 19.19 | — | — | — | — | — | — | — | |
| MAFParams=1.5M2026.06 | 19.07 | 15.62 | 15.65 | — | — | — | — | — | — | — | |
| VisualBERTParams=123M2026.06 | 18.17 | 6.08 | 8.06 | 69.01 | 36.08 | 41.03 | 21.25 | 57.1 | 28.67 | 25.92 | |
| UNITERParams=303M2026.06 | 16.92 | 7.15 | 8.83 | 68.34 | 44.29 | 50.22 | 28.12 | 72.78 | 38.91 | 32.65 | |
| VinVLParams=123M2026.06 | 16.76 | 8.6 | 9.75 | 68.49 | 62.32 | 61.3 | 42.88 | 80.81 | 53.69 | 41.58 | |
| LongdocParams=370M2026.06 | 7.79 | 8.43 | 7.24 | 62.27 | 76.1 | 67.69 | 48.77 | 84.95 | 61.02 | 45.31 | |
| Rule-BasedParams=N/A2026.06 | 5.6 | 10.13 | 6.4 | — | — | — | — | — | — | — | |
| LLaVA-OVParams=8B2026.06 | 0.2 | 17.59 | 11.17 | 75.83 | 94.88 | 83.75 | 77.21 | 82.38 | 78.8 | 57.91 | |
| Neural CorefParams=10M2026.06 | 0.11 | 0.17 | 0.13 | — | — | — | — | — | — | — |