Workshop Summary

This workshop is motivated by a fact: human beings have strong embodied intelligence in an open environment, but it is still challenging for large language models and LLM agents. Depsite some progresses on embodied AI on static and indoor environment, the LLM agents are still struggling in tasks in large-scale outdoor environment, such as navigation, search, spatial reasoning, task planning, etc. Therefore, we propose this workshop to discuss the recent advances on the related research area and looking forward to the future development. Specifically, it delves into topics of outdoor embodied intelligence, such as spatial intelligence and embodied perception, reasoning and planning, decision-making and action, multi-agent and human-agent collaboration, and the development of simulators, testbeds, datasets, and benchmarks. This comprehensive exploration of embodied LLM agents in open city environment holds the potential to advance the field of artificial intelligence and open up new applications in various domains.We also have a special poster/short paper session for those solutions that perform best in the Open Urban Environment Embodied Intelligence Competition.

We would like to discuss the following topics in this workshop:

(1) Spatial Intelligence and Embodied Perception with LLM Agents in Open City Environment:

1. How LLM agents can develop a sense of space and time in open city environments.
2. The role of embodied perception in enhancing the performance of LLM agents in outdoor environment.
3. Techniques for integrating spatial intelligence and embodied perception for LLM agents in outdoor environment.
4. Other related topics.

(2) Reasoning and planning with LLM agents in open city environment:

1. How LLM agents can use reasoning to make decisions in open city environment.
2. Strategies for planning actions and sequences of tasks for LLM agents in city environment.
3. Analysis on the bias and limitations of reasoning and planning of LLM.
4. Other related topics.

(3) Decision-making and Action with LLM agents in open city environment:

1. How LLM agents can make decisions based on outdoor context and goals.
2. Combination of large language models and small machine learning models for decision-making in outdoor environment.
3. Techniques for evaluating and improving the decision-making and action capabilities of LLM agents in outdoor environment.
4. Other related topics.

(4) Multi-agent and human-agent collaboration in open environment:

1. How multiple LLM agents can collaborate to achieve common goals in outdoor environment.
2. The challenges and opportunities of human-agent collaboration in open city environment.
3. Strategies for designing effective multi-agent systems in open city environment
4. Perspectives on human-AI system for outdoor applications.
5. Other related topics.

(5) Simulator, testbeds, datasets, benchmark for embodied LLM agent in city environment:

1. The development and use of simulators and testbeds for evaluating embodied LLM agents in outdoor environment.
2. The creation and curation of datasets for training and testing embodied LLM agents in outdoor environment.
3. The establishment of benchmarks and evaluation metrics for embodied LLM agents in outdoor environment.
4. Other related topics.

Call for Paper

The workshop is seeking submissions of research and industrial papers, including those presenting initial research findings, visionary ideas, or comprehensive studies. Submissions should adhere to the ICLR proceedings format and fit into one of the following categories:

Key Dates

1. Submission Deadline: Mar 03 2025 11:59PM UTC-0 Delayed. Please check the openreview site for the final deadline.

Submission Guidelines

1. - Format: Use the ICLR 2025 paper template. Ensure the appendix is part of the same PDF as the main document.
2. - Submit your paper via OpenReview.

Tentative schedule

Organizers and Biographies

Chen Gao

(Tsinghua University)

Yitao Liang

(Peking Univesity)

Xin Wang

(Tsinghua University)

Yu Zheng

(MIT)

Tong Xia

(University of Cambridge)

Fengli Xu

(Tsinghua University)

Yong Li

(Tsinghua University)