Research map / Multi-agent systems and games / LLM agents
Sentinel: Embodied Cooperative Spatial Reasoning and Planning
Xiangye Lin, Hongxin Zhang, Ruxi Deng, Qinhong Zhou, Chuang Gan
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 |
|---|---|---|---|
| add route refinement (ablation) | CoSaR · Sentinel 5 agents, 10 stationary sentinels | success rate (%): 26.19 -> 32.14 | Improved |
| add message analyzer (ablation) | CoSaR · Sentinel 5 agents, 10 stationary sentinels | success rate (%): 19.26 -> 32.14 | Improved |
| add spatial memory (ablation) | CoSaR · Sentinel 5 agents, 10 stationary sentinels | success rate (%): 21.42 -> 32.14 | Improved |
| add emergency avoidance (ablation) | CoSaR · Sentinel 5 agents, 10 stationary sentinels | success rate (%): 24.05 -> 32.14 | Improved |
| add emergency avoidance (ablation) | CoSaR · Sentinel 5 agents, 10 stationary sentinels | caught rate (%): 31.95 -> 21.43 | Improved |
| add message analyzer (ablation) | CoSaR · Sentinel 5 agents, 10 stationary sentinels | caught rate (%): 31.38 -> 21.43 | Improved |
| use gpt-4o backbone instead of Qwen3-VL-30B (ablation) | CoSaR · Sentinel 5 agents, 5 stationary sentinels | success rate (%): 60.71 -> 67.85 | Improved |
| add spatial-aware reasoning and memory (CoSaR instead of CoELA) | CoELA · Sentinel 5 agents, 10 stationary sentinels | success rate (%): 16.67 -> 32.14 | Improved |
| provide oracle perception | CoSaR · Sentinel 5 agents, 10 stationary sentinels | success rate (%): 32.14 -> 53.57 | Improved |
| increase sentinels from 5 to 20 (patrolling, 5 agents) | CoSaR · Sentinel patrolling | success rate (%): 39.29 -> 21.43 | Hurt |
| switch sentinels from stationary to patrolling | CoSaR · Sentinel 5 agents, 10 stationary sentinels | caught rate (%): 21.43 -> 27.62 | Hurt |
Closest papers by meaning
- Collaborative Tree Search for Enhancing Embodied Multi-Agent Collaboration (CVPR 2025 · similarity 0.86)
- From Assumptions to Actions: Turning LLM Reasoning into Uncertainty-Aware Planning for Embodied Agents (ICLR 2026 · similarity 0.86)
- SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning (NeurIPS 2026 · similarity 0.86)
- Mosaic: Runtime-Efficient Multi-Agent Embodied Planning (ICML 2026 · similarity 0.85)
- LLawCo: Learning Laws of Cooperation for Modeling Embodied Multi-Agent Behavior (ICML 2026 · similarity 0.85)
- CaPo: Cooperative Plan Optimization for Efficient Embodied Multi-Agent Cooperation (ICLR 2025 · similarity 0.85)
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.
