Research map / Large language models / Agents, code and math
Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Apply semantic action translation, trajectory correction, and dynamic example selection | GPT-3 175B planner · VirtualHome | Executability: 7.79% -> 73.05% | Helped |
| Apply semantic action translation, trajectory correction, and dynamic example selection | GPT-3 175B planner · VirtualHome | Correctness: 77.86% (6.42%) -> 66.13% (8.38%) | Hurt |
| Apply semantic action translation, trajectory correction, and dynamic example selection | GPT-3 175B planner · VirtualHome | LCS: 17.82% -> 24.09% | Helped |
| Add semantic action translation, trajectory correction, and dynamic example selection | Codex 12B planner · VirtualHome | Executability: 18.07% -> 78.57% | Helped |
| Add semantic action translation, trajectory correction, and dynamic example selection | Codex 12B planner · VirtualHome | Correctness: 64.87% (5.41%) -> 54.88% (5.90%) | Hurt |
| Add semantic action translation, trajectory correction, and dynamic example selection | Codex 12B planner · VirtualHome | LCS: 16.97% -> 24.72% | Helped |
| Add action translation (ablation) | Translated Codex 12B · VirtualHome | Executability: 31.49% -> 78.57% | Helped |
| Add action translation (ablation) | Translated GPT-3 175B · VirtualHome | Executability: 36.04% -> 73.05% | Helped |
| Add dynamic example selection (ablation) | Translated Codex 12B · VirtualHome | Executability: 50.86% -> 78.57% | Helped |
| Add dynamic example selection (ablation) | Translated GPT-3 175B · VirtualHome | Executability: 60.82% -> 73.05% | Helped |
| Add trajectory correction (ablation) | Translated Codex 12B · VirtualHome | Executability: 55.19% -> 78.57% | Helped |
| Add trajectory correction (ablation) | Translated GPT-3 175B · VirtualHome | LCS: 24.98% -> 24.09% | Hurt |
Closest papers by meaning
- R2C: Mapping Room to Chessboard to Unlock LLM As Low-Level Action Planner (CVPR 2025 · similarity 0.89)
- Agent Planning with World Knowledge Model (NeurIPS 2024 · similarity 0.89)
- Embodied Task Planning via Graph-Informed Action Generation with Large Language Models (ICML 2026 · similarity 0.89)
- Multi-Modal Grounded Planning and Efficient Replanning for Learning Embodied Agents with a Few Examples (AAAI 2025 · similarity 0.89)
- LoTa-Bench: Benchmarking Language-oriented Task Planners for Embodied Agents (ICLR 2024 · similarity 0.89)
- Embodied CoT Distillation From LLM To Off-the-shelf Agents (ICML 2024 · similarity 0.89)
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.