Survival on G2U Survival
0MeanRandom
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| RandomParadigm=Control Group2025.11 | 0 | 16.2 | 6.7 | |
| GPT-4oParadigm=Multimodal Large Language Model (MLLM) Group, Optimization Objective (@)=curiosity2025.11 | 0.079 | 16.8 | 11.6 | |
| Qwen3-VL-30B-A3BParadigm=Multimodal Large Language Model (MLLM) Group, Optimization Objective (@)=curiosity2025.11 | 0.086 | 15.8 | 11.6 | |
| GPT-4oParadigm=Multimodal Large Language Model (MLLM) Group, Optimization Objective (@)=utility2025.11 | 0.087 | 15.8 | 10.4 | |
| ACT-BCParadigm=Imitation Learning (IL) Group2025.11 | 0.088 | 14.3 | 17.1 | |
| Qwen3-VL-30B-A3BParadigm=Multimodal Large Language Model (MLLM) Group, Optimization Objective (@)=survival2025.11 | 0.091 | 14.3 | 11 | |
| GPT-5Paradigm=Multimodal Large Language Model (MLLM) Group, Optimization Objective (@)=curiosity2025.11 | 0.093 | 15.3 | 12.2 | |
| DreamerV3Paradigm=World Model Group, Optimization Objective (@)=utility2025.11 | 0.097 | 14.9 | 17.7 | |
| Qwen3-VL-8B-BCParadigm=Imitation Learning (IL) Group2025.11 | 0.099 | 12.9 | 14 | |
| V-JEPA2Paradigm=World Model Group, Optimization Objective (@)=curiosity2025.11 | 0.1 | 17.8 | 2.4 | |
| DreamerV3Paradigm=World Model Group, Optimization Objective (@)=survival2025.11 | 0.102 | 15.8 | 15.2 | |
| V-JEPA2Paradigm=World Model Group, Optimization Objective (@)=survival2025.11 | 0.102 | 17.4 | 4.9 | |
| V-JEPA2Paradigm=World Model Group, Optimization Objective (@)=utility2025.11 | 0.102 | 17.5 | 1.8 | |
| GenieReduxParadigm=World Model Group, Optimization Objective (@)=curiosity2025.11 | 0.104 | 14.3 | 14 | |
| Qwen3-VL-8BParadigm=Interactive Physical Reasoner Group, Interactive Physical Reasoner (IPR) Status=w/o IPR2025.11 | 0.105 | 13.7 | 14 | |
| DreamerV3Paradigm=World Model Group, Optimization Objective (@)=curiosity2025.11 | 0.108 | 14.5 | 17.7 | |
| GenieReduxParadigm=World Model Group, Optimization Objective (@)=survival2025.11 | 0.108 | 13.7 | 15.9 | |
| GPT-4oParadigm=Multimodal Large Language Model (MLLM) Group, Optimization Objective (@)=survival2025.11 | 0.108 | 12.6 | 13.4 | |
| GPT-5Paradigm=Multimodal Large Language Model (MLLM) Group, Optimization Objective (@)=utility2025.11 | 0.108 | 15.2 | 11 | |
| Qwen3-VL-30B-A3BParadigm=Multimodal Large Language Model (MLLM) Group, Optimization Objective (@)=utility2025.11 | 0.108 | 13.5 | 12.2 | |
| GenieReduxParadigm=World Model Group, Optimization Objective (@)=utility2025.11 | 0.11 | 13.7 | 16.5 | |
| PPOParadigm=Reinforcement Learning (RL) Group, Optimization Objective (@)=curiosity2025.11 | 0.114 | 14.9 | 11.6 | |
| PPOParadigm=Reinforcement Learning (RL) Group, Optimization Objective (@)=utility2025.11 | 0.12 | 15 | 12.2 | |
| DQNParadigm=Reinforcement Learning (RL) Group, Optimization Objective (@)=survival2025.11 | 0.121 | 14.4 | 15.9 | |
| PPOParadigm=Reinforcement Learning (RL) Group, Optimization Objective (@)=survival2025.11 | 0.125 | 14 | 14 | |
| DQNParadigm=Reinforcement Learning (RL) Group, Optimization Objective (@)=utility2025.11 | 0.125 | 13.7 | 16.5 | |
| DQNParadigm=Reinforcement Learning (RL) Group, Optimization Objective (@)=curiosity2025.11 | 0.131 | 13.2 | 18.3 | |
| GPT-5Paradigm=Multimodal Large Language Model (MLLM) Group, Optimization Objective (@)=survival2025.11 | 0.14 | 10.5 | 24.4 | |
| Qwen3-VL-8BParadigm=Interactive Physical Reasoner Group, Interactive Physical Reasoner (IPR) Status=w/ IPR2025.11 | 0.252 | 2.6 | 72 | |
| HumanParadigm=Control Group2025.11 | 1 | 5.7 | 46.3 |