Proactive Action Prediction on ABCD+
0.7293PRIQwen2.5-14B-ProActor-Q4
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Qwen2.5-14B-ProActor-Q4RULER=Custom2026.05 | 0.7293 | 0.426 | 0.484 | 0.136 | 0.2347 | 7.08 | 54.6 | |
| Qwen2.5-14B-ProActor-Q4RULER=Custom2026.05 | 0.7293 | 0.426 | 0.484 | 0.136 | 0.2347 | 7.08 | 54.6 | |
| Qwen2.5-14B-ProActor-Q4RULER=Adaptive2026.05 | 0.6842 | 0.431 | 0.586 | 0.32 | 0.2515 | 10.89 | 52.1 | |
| Qwen2.5-14B-ProActor-Q4RULER=Adaptive2026.05 | 0.6842 | 0.431 | 0.586 | 0.32 | 0.2515 | 10.89 | 52.1 | |
| Claude-4Reasoning=true2026.05 | 0.6318 | 0.421 | 0.749 | 0.779 | 0.2136 | 4.82 | 31.4 | |
| Claude-4Reasoning=true, ASG=false2026.05 | 0.6318 | 0.421 | 0.749 | 0.779 | 0.2136 | 4.82 | 31.4 | |
| Gemini-2.5-flashReasoning=false2026.05 | 0.6251 | 0.417 | 0.834 | 1 | 0.2133 | 3.99 | 28.8 | |
| Gemini-2.5-flashReasoning=false, ASG=false2026.05 | 0.6251 | 0.417 | 0.834 | 1 | 0.2133 | 3.99 | 28.8 | |
| Qwen2.5-14B-InstructReasoning=true2026.05 | 0.6246 | 0.423 | 0.449 | 0.062 | 0.1842 | 3.07 | 27.9 | |
| Qwen2.5-14B-InstructReasoning=true, ASG=false2026.05 | 0.6246 | 0.423 | 0.449 | 0.062 | 0.1842 | 3.07 | 27.9 | |
| Gemini-2.5-flashReasoning=true2026.05 | 0.6216 | 0.43 | 0.813 | 0.891 | 0.1782 | 2.45 | 25.2 | |
| Claude-4Reasoning=false2026.05 | 0.6216 | 0.427 | 0.831 | 0.946 | 0.2119 | 5.26 | 29.7 | |
| Gemini-2.5-flashReasoning=true, ASG=false2026.05 | 0.6216 | 0.43 | 0.813 | 0.891 | 0.1782 | 2.45 | 25.2 | |
| Claude-4Reasoning=false, ASG=false2026.05 | 0.6216 | 0.427 | 0.831 | 0.946 | 0.2119 | 5.26 | 29.7 | |
| Claude-4Reasoning=true, ASG=true2026.05 | 0.6031 | 0.403 | 0.852 | 1.114 | 0.2269 | 6.34 | 36 | |
| Claude-4Reasoning=true, ASG=true2026.05 | 0.6031 | 0.403 | 0.852 | 1.114 | 0.2269 | 6.34 | 36 | |
| GPT-5.1Reasoning=true2026.05 | 0.6003 | 0.429 | 0.789 | 0.839 | 0.1643 | 1.65 | 21.4 | |
| GPT-5.1Reasoning=true, ASG=false2026.05 | 0.6003 | 0.429 | 0.789 | 0.839 | 0.1643 | 1.65 | 21.4 | |
| Qwen2.5-14B-Instructmode=SFT (Full-tune)2026.05 | 0.5553 | 0.482 | 0.486 | 0.008 | 0.1141 | 0.34 | 13 | |
| GPT-5.1Reasoning=true, ASG=true2026.05 | 0.5547 | 0.42 | 0.733 | 0.712 | 0.1874 | 6.47 | 28.1 | |
| GPT-5.1Reasoning=true, ASG=true2026.05 | 0.5547 | 0.42 | 0.733 | 0.712 | 0.1874 | 6.47 | 28.1 | |
| Gemini-2.5-flashReasoning=true, ASG=true2026.05 | 0.5257 | 0.384 | 0.737 | 0.919 | 0.1715 | 2.86 | 24.1 | |
| Gemini-2.5-flashReasoning=true, ASG=true2026.05 | 0.5257 | 0.384 | 0.737 | 0.919 | 0.1715 | 2.86 | 24.1 | |
| GPT-5.1Reasoning=false2026.05 | 0.5104 | 0.318 | 0.706 | 0.717 | 0.2023 | 4.6 | 41.9 | |
| GPT-5.1Reasoning=false, ASG=false2026.05 | 0.5104 | 0.318 | 0.706 | 0.717 | 0.2023 | 4.6 | 41.9 | |
| GPT-4.1-miniReasoning=false2026.05 | 0.4405 | 0.363 | 0.524 | 0.363 | 0.0989 | 0.8 | 16.1 | |
| Qwen2.5-14B-InstructReasoning=true, ASG=true2026.05 | 0.4331 | 0.316 | 0.385 | 0.218 | 0.1918 | 4.32 | 35.9 | |
| Qwen2.5-14B-InstructReasoning=true, ASG=true2026.05 | 0.4331 | 0.316 | 0.385 | 0.218 | 0.1918 | 4.32 | 35.9 | |
| GPT-4.1-miniReasoning=true, ASG=true2026.05 | 0.4092 | 0.303 | 0.537 | 0.303 | 0.0625 | 0.67 | 28.1 | |
| GPT-4.1-miniReasoning=true2026.05 | 0.3986 | 0.315 | 0.55 | 0.315 | 0.0587 | 0.44 | 21.4 | |
| Qwen2.5-14B-InstructReasoning=false2026.05 | 0.2996 | 0.295 | 0.564 | 0.914 | 0.2237 | 7.73 | 51.5 | |
| Qwen2.5-14B-InstructReasoning=false, ASG=false2026.05 | 0.2996 | 0.295 | 0.564 | 0.914 | 0.2237 | 7.73 | 51.5 | |
| Qwen2.5-14B-Instructmode=SFT (LoRA)2026.05 | 0.17 | 0.272 | 0.533 | 0.96 | 0.2097 | 9.12 | 53.1 | |
| Qwen2.5-14B-InstructSFT=true2026.05 | 0.17 | 0.272 | 0.533 | 0.96 | 0.2097 | 9.12 | 53.1 |