TY - GEN
T1 - TPE
T2 - 13th CCF International Conference on Natural Language Processing and Chinese Computing, NLPCC 2024
AU - Wang, Hongru
AU - Wang, Huimin
AU - Wang, Lingzhi
AU - Hu, Minda
AU - Wang, Rui
AU - Xue, Boyang
AU - Huang, Yongfeng
AU - Wong, Kam Fai
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.
PY - 2025
Y1 - 2025
N2 - Previous works in tool learning mainly focus on the external function tools, such as models and APIs, while overlooking the existence of cognitive tools inside the cognitive/thinking processing of human mind. In this paper, we expand the scope of tools, centering on cognitive tools within the context of dialogue systems, such as multiple sources and psychological or tutoring strategies being dynamically applied in a single turn to compose helpful responses. To further enhance the reasoning and planning capability of LLMs with these cognitive tools, we introduce a multi-persona collaboration framework: Think-Plan-Execute (TPE). This framework decouples the response generation process into three distinct roles: Thinker, Planner, and Executor. Specifically, the Thinker analyzes the internal status exhibited in the dialogue context, such as user emotions and preferences, to formulate a global guideline. The Planner then generates executable plans to call different cognitive tools (e.g., sources or strategies), while the Executor compiles all intermediate results into a coherent response. Experimental results on three public datasets (FoCus, CIMA, and PsyQA) demonstrate the effectiveness and robustness of TPE, revealing the promising potential of cognitive tools as a substantial expansion of current toolset.
AB - Previous works in tool learning mainly focus on the external function tools, such as models and APIs, while overlooking the existence of cognitive tools inside the cognitive/thinking processing of human mind. In this paper, we expand the scope of tools, centering on cognitive tools within the context of dialogue systems, such as multiple sources and psychological or tutoring strategies being dynamically applied in a single turn to compose helpful responses. To further enhance the reasoning and planning capability of LLMs with these cognitive tools, we introduce a multi-persona collaboration framework: Think-Plan-Execute (TPE). This framework decouples the response generation process into three distinct roles: Thinker, Planner, and Executor. Specifically, the Thinker analyzes the internal status exhibited in the dialogue context, such as user emotions and preferences, to formulate a global guideline. The Planner then generates executable plans to call different cognitive tools (e.g., sources or strategies), while the Executor compiles all intermediate results into a coherent response. Experimental results on three public datasets (FoCus, CIMA, and PsyQA) demonstrate the effectiveness and robustness of TPE, revealing the promising potential of cognitive tools as a substantial expansion of current toolset.
KW - Dialogue System
KW - Large Language Model
KW - Tool Learning
UR - https://www.scopus.com/pages/publications/85210099092
U2 - 10.1007/978-981-97-9434-8_22
DO - 10.1007/978-981-97-9434-8_22
M3 - 会议稿件
AN - SCOPUS:85210099092
SN - 9789819794331
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 281
EP - 294
BT - Natural Language Processing and Chinese Computing - 13th National CCF Conference, NLPCC 2024, Proceedings
A2 - Wong, Derek F.
A2 - Wei, Zhongyu
A2 - Yang, Muyun
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 1 November 2024 through 3 November 2024
ER -