Position: Towards Unified Alignment Between Agents, Humans, and Environment

Zonghan Yang, An Liu, Zijun Liu, Kaiming Liu, Fangzhou Xiong, Yile Wang, Zeyuan Yang, Qingyuan Hu, Xinrui Chen, Zhenhe Zhang, Fuwen Luo, Zhicheng Guo, Peng Li, Yang Liu
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:56251-56275, 2024.

Abstract

The rapid progress of foundation models has led to the prosperity of autonomous agents, which leverage the universal capabilities of foundation models to conduct reasoning, decision-making, and environmental interaction. However, the efficacy of agents remains limited when operating in intricate, realistic environments. In this work, we introduce the principles of Unified Alignment for Agents (UA$^2$), which advocate for the simultaneous alignment of agents with human intentions, environmental dynamics, and self-constraints such as the limitation of monetary budgets. From the perspective of UA$^2$, we review the current agent research and highlight the neglected factors in existing agent benchmarks and method candidates. We also conduct proof-of-concept studies by introducing realistic features to WebShop, including user profiles demonstrating intentions, personalized reranking reflecting complex environmental dynamics, and runtime cost statistics as self-constraints. We then follow the principles of UA$^2$ to propose an initial design of our agent and benchmark its performance with several candidate baselines in the retrofitted WebShop. The extensive experimental results further prove the importance of the principles of UA$^2$. Our research sheds light on the next steps of autonomous agent research with improved general problem-solving abilities.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-yang24p, title = {Position: Towards Unified Alignment Between Agents, Humans, and Environment}, author = {Yang, Zonghan and Liu, An and Liu, Zijun and Liu, Kaiming and Xiong, Fangzhou and Wang, Yile and Yang, Zeyuan and Hu, Qingyuan and Chen, Xinrui and Zhang, Zhenhe and Luo, Fuwen and Guo, Zhicheng and Li, Peng and Liu, Yang}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {56251--56275}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24p/yang24p.pdf}, url = {https://proceedings.mlr.press/v235/yang24p.html}, abstract = {The rapid progress of foundation models has led to the prosperity of autonomous agents, which leverage the universal capabilities of foundation models to conduct reasoning, decision-making, and environmental interaction. However, the efficacy of agents remains limited when operating in intricate, realistic environments. In this work, we introduce the principles of Unified Alignment for Agents (UA$^2$), which advocate for the simultaneous alignment of agents with human intentions, environmental dynamics, and self-constraints such as the limitation of monetary budgets. From the perspective of UA$^2$, we review the current agent research and highlight the neglected factors in existing agent benchmarks and method candidates. We also conduct proof-of-concept studies by introducing realistic features to WebShop, including user profiles demonstrating intentions, personalized reranking reflecting complex environmental dynamics, and runtime cost statistics as self-constraints. We then follow the principles of UA$^2$ to propose an initial design of our agent and benchmark its performance with several candidate baselines in the retrofitted WebShop. The extensive experimental results further prove the importance of the principles of UA$^2$. Our research sheds light on the next steps of autonomous agent research with improved general problem-solving abilities.} }
Endnote
%0 Conference Paper %T Position: Towards Unified Alignment Between Agents, Humans, and Environment %A Zonghan Yang %A An Liu %A Zijun Liu %A Kaiming Liu %A Fangzhou Xiong %A Yile Wang %A Zeyuan Yang %A Qingyuan Hu %A Xinrui Chen %A Zhenhe Zhang %A Fuwen Luo %A Zhicheng Guo %A Peng Li %A Yang Liu %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-yang24p %I PMLR %P 56251--56275 %U https://proceedings.mlr.press/v235/yang24p.html %V 235 %X The rapid progress of foundation models has led to the prosperity of autonomous agents, which leverage the universal capabilities of foundation models to conduct reasoning, decision-making, and environmental interaction. However, the efficacy of agents remains limited when operating in intricate, realistic environments. In this work, we introduce the principles of Unified Alignment for Agents (UA$^2$), which advocate for the simultaneous alignment of agents with human intentions, environmental dynamics, and self-constraints such as the limitation of monetary budgets. From the perspective of UA$^2$, we review the current agent research and highlight the neglected factors in existing agent benchmarks and method candidates. We also conduct proof-of-concept studies by introducing realistic features to WebShop, including user profiles demonstrating intentions, personalized reranking reflecting complex environmental dynamics, and runtime cost statistics as self-constraints. We then follow the principles of UA$^2$ to propose an initial design of our agent and benchmark its performance with several candidate baselines in the retrofitted WebShop. The extensive experimental results further prove the importance of the principles of UA$^2$. Our research sheds light on the next steps of autonomous agent research with improved general problem-solving abilities.
APA
Yang, Z., Liu, A., Liu, Z., Liu, K., Xiong, F., Wang, Y., Yang, Z., Hu, Q., Chen, X., Zhang, Z., Luo, F., Guo, Z., Li, P. & Liu, Y.. (2024). Position: Towards Unified Alignment Between Agents, Humans, and Environment. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:56251-56275 Available from https://proceedings.mlr.press/v235/yang24p.html.

Related Material