Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Robust Execution of BDI Agent Programs by Exploiting Synergies Between Intentions
Yuan Yao, Brian Logan, John Thangarajah
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 |
|---|---|---|---|
| replace Round Robin with failure handling (RR+) by SP scheduling exploiting intention synergies | BDI agent · Synthetic goal-plan trees, |V|=50, 0% fallible | backtracks: 13.82 -> 0 | Improved |
| use SP instead of RR+ | BDI agent · Non-disruptive failures, |V|=20, 30% fallible, 50% failure prob | backtracks: 44.78 -> 1.76 | Improved |
| use SP instead of RR+ | BDI agent · Non-disruptive failures, |V|=50, 30% fallible, 50% failure prob | goals achieved (of 10): 7.62 -> 10 | Improved |
| use SP instead of RR+ | BDI agent · Disruptive failures, |V|=20, 30% fallible, 50% failure prob | backtracks: 36.92 -> 0.46 | Improved |
| use SP instead of RR+ with only five parallel intentions | BDI agent · 5 goal-plan trees, |V|=20, 30% non-disruptive | backtracks: 15.86 -> 2.38 | Improved |
| reduce SP iterations from 1000 to 100 (ablation) | SP · Non-disruptive, |V|=20, 30% fallible, 50% failure prob | goals achieved (of 10): 10 -> 9.98 | No clear effect |
| reduce SP iterations from 1000 to 100 (ablation) | SP · Non-disruptive, |V|=20, 30% fallible, 50% failure prob | backtracks: 1.76 -> 4.18 | Hurt |
| make fallible action failures disruptive instead of non-disruptive | RR+ · |V|=50, 30% fallible, 50% failure prob | backtracks: 30.96 -> 20.62 | Improved |
| increase interaction by lowering |V| from 50 to 20 | RR+ · 0% fallible | goals achieved (of 10): 9.32 -> 8.26 | Hurt |
Closest papers by meaning
- Agent-Oriented Planning in Multi-Agent Systems (ICLR 2025 · similarity 0.83)
- Bidirectional Temporal Plan Graph: Enabling Switchable Passing Orders for More Efficient Multi-Agent Path Finding Plan Execution (AAAI 2024 · similarity 0.82)
- PRP Rebooted: Advancing the State of the Art in FOND Planning (AAAI 2024 · similarity 0.81)
- On the Resilience of LLM-Based Multi-Agent Collaboration with Faulty Agents (ICML 2025 · similarity 0.81)
- EvoCodeBench: Evaluating Coding Agents in Multi-Turn Iterative Interactions (NeurIPS 2026 · similarity 0.81)
- EnCompass: Enhancing Agent Programming with Search Over Program Execution Paths (NeurIPS 2025 · similarity 0.81)
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.
