k4ng
/

SCOPE-RL-9B

Image-Text-to-Text

computer-use-agent

reinforcement-learning

Model card Files Files and versions

k4ng commited on 13 days ago

Commit

6883941

·

verified ·

1 Parent(s): bf29e6a

Update README.md

Files changed (1) hide show

README.md +1 -2

README.md CHANGED Viewed

@@ -68,9 +68,8 @@ vllm serve k4ng/SCOPE-RL-9B \
 | Open-source | ComputerRL | 21.77 | 48.90 | 14.00 |
 | Open-source | Qwen3.5-9B | 8.93 | 41.80 | 5.00 |
 | Open-source | Qwen3-VL-8B | 16.27 | 33.90 | 10.70 |
-| SCOPE | SCOPE-mid-checkpoint | 48.64 | 45.43 | 52.33 |
 | SCOPE | SCOPE-Capability-Safety | 56.83 | 49.72 | **66.30** |
-| **SCOPE** | **SCOPE-RL-9B** | **58.80** | **54.17** | 64.30 |
 All values are percentages, and ↑ indicates that higher is better. \(H\) is the harmonic mean of the OSWorld task success rate and the OS-BLIND attack-avoidance rate. Compared with SCOPE-Capability-Safety, SCOPE-RL-9B improves OSWorld performance by 4.45 percentage points, while attack avoidance decreases by 2.00 points. The harmonic mean increases from 56.83% to 58.80%.

 | Open-source | ComputerRL | 21.77 | 48.90 | 14.00 |
 | Open-source | Qwen3.5-9B | 8.93 | 41.80 | 5.00 |
 | Open-source | Qwen3-VL-8B | 16.27 | 33.90 | 10.70 |
 | SCOPE | SCOPE-Capability-Safety | 56.83 | 49.72 | **66.30** |
+| SCOPE | **SCOPE-RL-9B** | **58.80** | **54.17** | 64.30 |
 All values are percentages, and ↑ indicates that higher is better. \(H\) is the harmonic mean of the OSWorld task success rate and the OS-BLIND attack-avoidance rate. Compared with SCOPE-Capability-Safety, SCOPE-RL-9B improves OSWorld performance by 4.45 percentage points, while attack avoidance decreases by 2.00 points. The harmonic mean increases from 56.83% to 58.80%.