Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Open-World Multi-Task Control Through Goal-Aware Representation Learning and Adaptive Horizon Prediction
Shaofei Cai, Zihao Wang, Xiaojian Ma, Anji Liu, Yitao Liang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use the full goal-sensitive backbone and adaptive horizon method instead of MT-BC | Minecraft multi-task control · Plains, Flat, and Forest | Avg. Success Rate and Avg. Precision: Plains: 32_{\pm05} -> \boldsymbol{55}_{\pm09} (success rate), 43_{\pm06} -> \boldsymbol{70}_{\pm09} (precision); Flat: 3 | Improved |
| add goal-sensitive backbone features (ablation) | Ours and MT-BC · Plains, Flat, and Forest multi-task benchmarks | Success Rate and Precision | Improved |
| add adaptive horizon prediction (ablation) | Ours · Plains multi-task | Avg. Success Rate: 35_{\pm08} -> \boldsymbol{55}_{\pm09} | Improved |
| add a GRU module while retaining horizon prediction (ablation) | Ours · Plains multi-task | Avg. Success Rate: 55_{\pm09} -> \boldsymbol{65}_{\pm07} | Improved |
| add horizon prediction alongside a GRU module (ablation) | Ours without horizon prediction + RNN · Plains multi-task | Avg. Success Rate: 39_{\pm08} -> \boldsymbol{65}_{\pm07} | Improved |
| add horizon loss (ablation) | Ours · Plains multi-task | Avg. Success Rate: 47_{\pm06} -> \boldsymbol{55}_{\pm09} | Improved |
| include extra observation input (ablation) | Ours · Plains multi-task | Avg. Success Rate: 50_{\pm07} -> \boldsymbol{55}_{\pm09} | Improved |
| include language condition (ablation) | Ours · Plains multi-task | Avg. Success Rate: 25_{\pm03} -> \boldsymbol{55}_{\pm09} | Improved |
| subtract a small positive constant from predicted horizon at evaluation (ablation) | Horizon-conditioned policy · Flat multi-task | Performance | Improved |
| subtract a larger value from predicted horizon (ablation) | Horizon-conditioned policy · Flat multi-task | Performance | Hurt |
| train on Plains and evaluate without fine-tuning on Flat | Ours · Flat novel-biome evaluation | Avg. Success Rate: 63 -> 58 | Mixed |
Closest papers by meaning
- MP5: A Multi-modal Open-ended Embodied System in Minecraft via Active Perception (CVPR 2024 · similarity 0.88)
- ADAM: An Embodied Causal Agent in Open-World Environments (ICLR 2025 · similarity 0.87)
- Open-World Reinforcement Learning over Long Short-Term Imagination (ICLR 2025 · similarity 0.87)
- GROOT: Learning to Follow Instructions by Watching Gameplay Videos (ICLR 2024 · similarity 0.87)
- Optimus-1: Hybrid Multimodal Memory Empowered Agents Excel in Long-Horizon Tasks (NeurIPS 2024 · similarity 0.86)
- ROCKET-1: Mastering Open-World Interaction with Visual-Temporal Context Prompting (CVPR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
