| Paper | Code | KRIS-Bench Overall | ModelName | ReleaseDate |
|---|---|---|---|---|
| Reasoning-Informed Visual Editing | ✓ Link | 88.12 | RISE-Agent | 2026-10-08 |
| OmniHarness: Harnessing Generalizable Visual Generation via Symbolic Policy Learning | ✓ Link | 77.33 | OmniHarness | 2026-09-13 |
| ThinkRL-Edit: Thinking in Reinforcement Learning for Reasoning-Centric Image Editing | 71.65 | ThinkRL-Edit | 2026-01-06 | |
| Uncovering Understanding-Generation Synergy in Native Unified Multimodal Models: From Representation, Task to System | 68.33 | Native UMM Synergy: End-to-end UMM | 2026-09-01 | |
| REASONEDIT: Towards Reasoning-Enhanced Image Editing Models | ✓ Link | 61.57 | ReasonEdit-Q (thinking + reflection) | 2025-11-27 |
| Emerging Properties in Unified Multimodal Pretraining | ✓ Link | 60.18 | BAGEL (7B active / 14B total MoT) with Self-CoT | 2025-05-20 |