Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Reducing Goal State Divergence with Environment Design
Kelsey Sikes, Sarah Keren, Sarath Sreedharan
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 planner-based compilation instead of naive breadth-first search over designs | Naive BFS · IPC Blocksworld | design time (s): 387.178 -> 73.849 | Improved |
| use planner-based compilation instead of naive breadth-first search over designs | Naive BFS · IPC Depot | design time (s): 188.556 -> 35.593 | Improved |
| use planner-based compilation instead of naive breadth-first search over designs | Naive BFS · IPC Elevator | design time (s): 17.930 -> 3.691 | Improved |
| use planner-based compilation instead of naive breadth-first search over designs | Naive BFS · IPC Logistics | design time (s): 50.306 -> 33.062 | Improved |
| use planner-based compilation instead of naive breadth-first search over designs | Naive BFS · IPC Zenotravel | design time (s): 23.641 -> 5.084 | Improved |
| enforce ordering between human and robot actions in the compilation (ablation) | Main-fl · IPC Blocksworld | design time (s): 73.172 -> 73.849 | No clear effect |
| enforce ordering between human and robot actions in the compilation (ablation) | Main-fl · IPC Depot | design time (s): 31.877 -> 35.593 | No clear effect |
| add design search to the lower-bound compilation | GD-lower bound · IPC Depot | time to compute bound (s): 5.747 -> 5.234 | No clear effect |
| compute the upper bound instead of the lower bound | GD-lower bound · IPC Logistics | time to compute bound (s): 0.886 -> 28.785 | Hurt |
Closest papers by meaning
- Goal Alignment: Re-analyzing Value Alignment Problems Using Human-Aware AI (AAAI 2024 · similarity 0.85)
- STAIR: Addressing Stage Misalignment through Temporal-Aligned Preference Reinforcement Learning (NeurIPS 2025 · similarity 0.85)
- Introspective Planning: Aligning Robots' Uncertainty with Inherent Task Ambiguity (NeurIPS 2024 · similarity 0.84)
- COOPERA: Continual Open-Ended Human-Robot Assistance (NeurIPS 2025 · similarity 0.84)
- AlignDiff: Aligning Diverse Human Preferences via Behavior-Customisable Diffusion Model (ICLR 2024 · similarity 0.84)
- DUO: Diverse, Uncertain, On-Policy Query Generation and Selection for Reinforcement Learning from Human Feedback (AAAI 2025 · similarity 0.83)
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.
