Vision Question Answering on WOD-E2E (val)
31BLEU-4MindVLA-U1 + Intent-CFG
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MindVLA-U1 + Intent-CFGBackbone=Qwen3-VL-2B, AD-PT=×2026.05 | 31 | 52 | |
| MindVLA-U1Backbone=Qwen3-VL-2B, AD-PT=×2026.05 | 30 | 49 | |
| MindVLA-U1 + MoTBackbone=Qwen3-VL-2B, AD-PT=×2026.05 | 30 | 51 |