General-Purpose Utility on General-Purpose Utility Evaluation v1 (avg)
92.3FluencyPrompting [52]
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Prompting [52]Target Model=Qwen2.5-VL-7B-Instruct [5], Steering Method=Prompting [52]2026.04 | 92.3 | 3.072 | |
| No SteeringTarget Model=Qwen2.5-VL-7B-Instruct [5], Steering Method=No Steering2026.04 | 91.7 | 3.027 | |
| No SteeringTarget Model=InternVL3.5-8B-Instruct [108], Steering Method=No Steering2026.04 | 91 | 2.984 | |
| No SteeringTarget Model=LLaVA1.6-Mistral-7B [48], Steering Method=No Steering2026.04 | 90.9 | 2.798 | |
| DACOTarget Model=Qwen2.5-VL-7B-Instruct [5], Steering Method=DACO (Ours)2026.04 | 90.5 | 3.137 | |
| DACOTarget Model=InternVL3.5-8B-Instruct [108], Steering Method=DACO (Ours)2026.04 | 89.8 | 3.005 | |
| Prompting [52]Target Model=InternVL3.5-8B-Instruct [108], Steering Method=Prompting [52]2026.04 | 89.5 | 2.947 | |
| Prompting [52]Target Model=LLaVA1.6-Mistral-7B [48], Steering Method=Prompting [52]2026.04 | 89.2 | 3.018 | |
| MOP [56]Target Model=InternVL3.5-8B-Instruct [108], Steering Method=MOP [56]2026.04 | 89 | 3.09 | |
| MOP [56]Target Model=LLaVA1.6-Mistral-7B [48], Steering Method=MOP [56]2026.04 | 88.7 | 2.998 | |
| DACOTarget Model=LLaVA1.6-Mistral-7B [48], Steering Method=DACO (Ours)2026.04 | 88 | 2.906 | |
| ActAdd [50, 101, 134]Target Model=LLaVA1.6-Mistral-7B [48], Steering Method=ActAdd [50, 101, 134]2026.04 | 85.8 | 3.251 | |
| MOP [56]Target Model=Qwen2.5-VL-7B-Instruct [5], Steering Method=MOP [56]2026.04 | 81.6 | 3.349 | |
| ActAdd [50, 101, 134]Target Model=InternVL3.5-8B-Instruct [108], Steering Method=ActAdd [50, 101, 134]2026.04 | 77.1 | 3.352 | |
| ActAdd [50, 101, 134]Target Model=Qwen2.5-VL-7B-Instruct [5], Steering Method=ActAdd [50, 101, 134]2026.04 | 69.1 | 3.818 |