Segmentation on MIMOSeg Perspective I held-in (test)
63.9mIoUMIMO
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| MIMOMulti-modal Input Aligner=true, LLM=Vicuna-7B2025.10 | 63.9 | 31.6 | 45.7 | |
| MIMO(w/o Aligner)Multi-modal Input Aligner=false, LLM=Vicuna-7B2025.10 | 60.7 | 30.1 | 44.8 | |
| GLaMMFine-tuned=true2025.10 | 55.6 | 32.4 | 50.7 |