--- title: "LLM-Planner: Few-Shot Grounded Planning for — Song 2022" description: "Language-driven embodied agents usually need many annotated instruction–trajectory pairs for every task, which makes them expensive to build and slow to…" authors: ["Chan Hee Song", "Jiaman Wu", "Clayton Washington", "Brian M. Sadler", "Wei-Lun Chao", "Yu Su"] year: 2022 venue: "arXiv:2212.04088" source: "https://arxiv.org/abs/2212.04088" doi: "10.48550/arxiv.2212.04088" arxiv: "2212.04088" licence: "CC BY 4.0 (arXiv)" licence_url: "http://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/song-2022-llm-planner-few-shot-grounded/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/song-2022-llm-planner-few-shot-grounded/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2022](https://md.research.elamur.ai/free/en/papers/2022/page/3/) / LLM-Planner: Few-Shot Grounded Planning for Embodied Agents with Large Language Models # LLM-Planner: Few-Shot Grounded Planning for Embodied Agents with Large Language Models **Authors:** Chan Hee Song, Jiaman Wu, Clayton Washington, Brian M. Sadler, Wei-Lun Chao, [Yu Su](https://md.research.elamur.ai/free/en/authors/yu-su/) · **Year:** 2022 · **Venue:** arXiv:2212.04088 · **Licence:** [CC BY 4.0 (arXiv)](http://creativecommons.org/licenses/by/4.0/) **Links:** [Publisher's page](https://arxiv.org/abs/2212.04088) · [PDF](https://arxiv.org/pdf/2212.04088) · [DOI 10.48550/arxiv.2212.04088](https://doi.org/10.48550/arxiv.2212.04088) · [arXiv 2212.04088](https://arxiv.org/abs/2212.04088) · [OpenAlex](https://openalex.org/W4390874280) **Citations:** 420 citations (openalex-snapshot · 2026-10-14) · [143](#cited-by) in our library, of the 140884 papers whose bibliographies we read ## Summary Language-driven embodied agents usually need many annotated instruction–trajectory pairs for every task, which makes them expensive to build and slow to extend. LLM-Planner addresses this by using a large language model (GPT-3) as the high-level planner in a hierarchical agent. Given an instruction and a few retrieved example pairs in the prompt, the LLM writes a sequence of subgoals such as navigating to or picking up an object. A low-level planner then turns each subgoal into primitive actions. When the agent stalls, taking too many steps or failing repeatedly, the LLM is prompted again with the subgoals already completed and the objects seen so far. It then produces a revised continuation that fits the actual environment. The planner is plugged into the perception and low-level modules of HLSM and tested on ALFRED. With only a tiny fraction of the paired data, it performs comparably to fully trained baselines, while other methods nearly fail in the same few-shot regime. This suggests a cheap route to versatile, sample-efficient agents. ## Key points - **Data efficiency:** The planner uses about 100 paired examples, under 0.5% of ALFRED's training data, and no parameter updates (in-context learning only). - **Prompting techniques:** The method adds a kNN retriever for dynamic in-context example selection and logit biases. In the ablation, LOOCV HLP accuracy is 40.59 for the full model. It falls to 17.48 without the retriever, 38.10 without logit biases, and 13.43 without both. - **Comparison with SayCan:** Unlike SayCan, which ranks a pre-enumerated list of admissible skills, LLM-Planner generates plans directly. This removes the need for prior environment knowledge and reduces the number of LLM calls. - **Hyperparameter sensitivity:** Around 9 in-context examples works well. Gains from more training examples level off near 250. - **Case studies:** Grounded re-planning helps locate objects, for example searching a cabinet for a missing cup. It also resolves ambiguous references, for example mapping "lamp" to the desklamp present in the scene. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* Fields: [Other](https://md.research.elamur.ai/free/en/shelves/other/) ## Cited by Cited by 143 papers of our library cite it (of the 140884 whose bibliographies we read); 10 of them have a page here, 133 not yet. - [From Persona to Personalization: A Survey on Role-Playing Language Agents](https://md.research.elamur.ai/free/en/papers/chen-2024-persona-personalization-survey-role-playing/) - [Mind2Web: Towards a Generalist Agent for the Web](https://md.research.elamur.ai/free/en/papers/deng-2023-mind2web-towards-generalist-agent-web/) - [Towards Reasoning in Large Language Models: A Survey](https://md.research.elamur.ai/free/en/papers/huang-2023-reasoning-large-language-models-survey/) - [A Comprehensive Overview of Large Language Models](https://md.research.elamur.ai/free/en/papers/naveed-2023-comprehensive-overview-large-language-models/) - [Tool Learning with Foundation Models](https://md.research.elamur.ai/free/en/papers/qin-2023-tool-learning-foundation-models/) - [UI-TARS: Pioneering Automated GUI Interaction with Native Agents](https://md.research.elamur.ai/free/en/papers/qin-2025-ui-tars-pioneering-automated/) - [Large Language Model Alignment: A Survey](https://md.research.elamur.ai/free/en/papers/shen-2023-large-language-model-alignment-survey/) - [基于大模型的具身智能系统综述](https://md.research.elamur.ai/free/en/papers/wang-2025-embodied-intelligence-systems/) - [The Rise and Potential of Large Language Model Based Agents: A Survey](https://md.research.elamur.ai/free/en/papers/xi-2023-rise-potential-large-language/) - [To Skip, to Swap or to Not Swap? Identifying Step Transition Types in Instructional Manuals](https://md.research.elamur.ai/free/en/papers/yang-2026-skip-swap-or-not-swap/) ## Cite The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @misc{song2022llm, title = {{LLM-Planner: Few-Shot Grounded Planning for Embodied Agents with Large Language Models}}, author = {Song, Chan Hee and Wu, Jiaman and Washington, Clayton and Sadler, Brian M. and Chao, Wei-Lun and Su, Yu}, year = {2022}, doi = {10.48550/arxiv.2212.04088}, eprint = {2212.04088}, archivePrefix = {arXiv}, url = {https://doi.org/10.48550/arxiv.2212.04088} } ``` - [← previous Lisan: Yemeni, Iraqi, Libyan, and Sudanese Arabic Dialect Copora with Morphological Annotations](https://md.research.elamur.ai/free/en/papers/jarrar-2022-lisan-yemeni-iraqi-libyan-sudanese/) - [next → LongT5: Efficient Text-To-Text Transformer for Long Sequences](https://md.research.elamur.ai/free/en/papers/guo-2022-longt5-efficient-text-text-transformer/) - [↑ back to papers of 2022](https://md.research.elamur.ai/free/en/papers/2022/page/3/) ## References 4 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Michael Ahn, Anthony Brohan, Noah Brown, Yevgen Chebotar, Omar Cortes, Byron David, Chelsea Finn, Chuyuan Fu, Keerthana Gopalakrishnan, Karol Hausman, Alex Herzog, Daniel Ho, Jasmine Hsu, Julian Ibarz, Brian Ichter, Alex Irpan, Eric Jang, Rosario Jauregui Ruano, Kyle Jeffrey, Sally Jesmonth, Nikhil… [find](https://scholar.google.com/scholar?q=Michael+Ahn%2C+Anthony+Brohan%2C+Noah+Brown%2C+Yevgen+Chebotar%2C+Omar+Cortes%2C+Byron+David%2C+Chelsea+Finn%2C+Chuyuan+Fu%2C+Keerthana+Gopalakrishnan%2C+Karol+Hausman%2C+Alex+Herzog%2C+Daniel+Ho%2C+Jasmine+Hsu%2C+Julian+Ibarz) [our card](https://md.research.elamur.ai/free/en/papers/ahn-2022-i-i-say-grounding-language/) 2. Peter Anderson, Qi Wu, Damien Teney, Jake Bruce, Mark Johnson, Niko Sunderhauf, Ian Reid, Stephen Gould, and Anton Van Den Hengel. Vision-and-language navigation: Interpreting visually-grounded navigation instructions in real environments. In CVPR, pages 3674–3683, 2018. 1, 3 [find](https://scholar.google.com/scholar?q=Peter+Anderson%2C+Qi+Wu%2C+Damien+Teney%2C+Jake+Bruce%2C+Mark+Johnson%2C+Niko+Sunderhauf%2C+Ian+Reid%2C+Stephen+Gould%2C+and+Anton+Van+Den+Hengel%2E+Vision%2Dand%2Dlanguage+navigation%3A+Interpreting+visually%2Dgrounded+naviga) [our card](https://md.research.elamur.ai/free/en/papers/anderson-2017-vision-language-navigation-interpreting-visually/) 3. Valts Blukis, Chris Paxton, Dieter Fox, Animesh Garg, and Yoav Artzi. A persistent spatial semantic representation for high-level natural language instruction execution. In Conference on Robot Learning, pages 706–717. PMLR, 2022. 2, 3, 4, 6, 8, 12, 14 [find](https://scholar.google.com/scholar?q=Valts+Blukis%2C+Chris+Paxton%2C+Dieter+Fox%2C+Animesh+Garg%2C+and+Yoav+Artzi%2E+A+persistent+spatial+semantic+representation+for+high%2Dlevel+natural+language+instruction+execution%2E+In+Conference+on+Robot+Learnin) 4. Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel Ziegler, Jeffrey Wu, Clemens Winter, Chri… [find](https://scholar.google.com/scholar?q=Tom+Brown%2C+Benjamin+Mann%2C+Nick+Ryder%2C+Melanie+Subbiah%2C+Jared+D+Kaplan%2C+Prafulla+Dhariwal%2C+Arvind+Neelakantan%2C+Pranav+Shyam%2C+Girish+Sastry%2C+Amanda+Askell%2C+Sandhini+Agarwal%2C+Ariel+Herbert%2DVoss%2C+Gretchen) [our card](https://md.research.elamur.ai/free/en/papers/huang-2022-language-models-zero-shot-planners/)