TPE: Towards Better Compositional Reasoning over Conceptual Tools with Multi-persona Collaboration (2309.16090v1)
Abstract: LLMs have demonstrated exceptional performance in planning the use of various functional tools, such as calculators and retrievers, particularly in question-answering tasks. In this paper, we expand the definition of these tools, centering on conceptual tools within the context of dialogue systems. A conceptual tool specifies a cognitive concept that aids systematic or investigative thought. These conceptual tools play important roles in practice, such as multiple 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 conceptual 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 conceptual tools (e.g., sources or strategies), while the Executor compiles all intermediate results into a coherent response. This structured approach not only enhances the explainability and controllability of responses but also reduces token redundancy. We demonstrate the effectiveness of TPE across various dialogue response generation tasks, including multi-source (FoCus) and multi-strategy interactions (CIMA and PsyQA). This reveals its potential to handle real-world dialogue interactions that require more complicated tool learning beyond just functional tools. The full code and data will be released for reproduction.
- A multitask, multilingual, multimodal evaluation of chatgpt on reasoning, hallucination, and interactivity. arXiv preprint arXiv:2302.04023, 2023.
- Plato-k: Internal and external knowledge enhanced dialogue generation. arXiv preprint arXiv:2211.00910, 2022.
- Simulation of networks of spiking neurons: a review of tools and strategies. Journal of computational neuroscience, 23:349–398, 2007.
- Language models are few-shot learners. Advances in neural information processing systems, 33:1877–1901, 2020.
- Agentverse: Facilitating multi-agent collaboration and exploring emergent behaviors in agents. arXiv preprint arXiv:2308.10848, 2023.
- Improving multi-turn emotional support dialogue generation with lookahead strategy planning. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pp. 3014–3026, Abu Dhabi, United Arab Emirates, December 2022. Association for Computational Linguistics. doi: 10.18653/v1/2022.emnlp-main.195. URL https://aclanthology.org/2022.emnlp-main.195.
- Wizard of wikipedia: Knowledge-powered conversational agents. In 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019. OpenReview.net, 2019. URL https://openreview.net/forum?id=r1l73iRqKm.
- Vanessa Dye. Reflection, reflection, reflection. i’m thinking all the time, why do i need a theory or model of reflection?’. Developing Reflective Practice: A guide for beginning teachers. Maidenhead: McGraw-Hill Education, pp. 217–234, 2011.
- Defensive tool use in a coconut-carrying octopus. Current biology, 19(23):R1069–R1070, 2009.
- There are a thousand hamlets in a thousand people’s eyes: Enhancing knowledge-grounded dialogue with personal memory. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 3901–3913, Dublin, Ireland, May 2022. Association for Computational Linguistics. doi: 10.18653/v1/2022.acl-long.270. URL https://aclanthology.org/2022.acl-long.270.
- Tools for Strategy: A Starter Kit for Academics and Practitioners. Elements in Business Strategy. Cambridge University Press, 2020. doi: 10.1017/9781108883757.
- Decoupling strategy and generation in negotiation dialogues. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pp. 2333–2343, Brussels, Belgium, October-November 2018. Association for Computational Linguistics. doi: 10.18653/v1/D18-1256. URL https://aclanthology.org/D18-1256.
- Clara E Hill. Helping skills: Facilitating, exploration, insight, and action. American Psychological Association, 2009.
- Tool documentation enables zero-shot tool-usage with large language models. arXiv preprint arXiv:2308.00675, 2023.
- Call for customized conversation: Customized conversation grounding persona and knowledge. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, pp. 10803–10812, 2022.
- Survey of hallucination in natural language generation. ACM Comput. Surv., 55(12), mar 2023. ISSN 0360-0300. doi: 10.1145/3571730. URL https://doi.org/10.1145/3571730.
- Tool-making and tool-using in the northern blue jay. Science, 180(4090):1076–1078, 1973.
- Dense passage retrieval for open-domain question answering, 2020. URL https://arxiv.org/abs/2004.04906.
- Internet-augmented dialogue generation. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 8460–8478, Dublin, Ireland, May 2022. Association for Computational Linguistics. doi: 10.18653/v1/2022.acl-long.579. URL https://aclanthology.org/2022.acl-long.579.
- Understanding client reactions in online mental health counseling. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 10358–10376, Toronto, Canada, July 2023a. Association for Computational Linguistics. doi: 10.18653/v1/2023.acl-long.577. URL https://aclanthology.org/2023.acl-long.577.
- Api-bank: A benchmark for tool-augmented llms, 2023b.
- Code as policies: Language model programs for embodied control. In 2023 IEEE International Conference on Robotics and Automation (ICRA), pp. 9493–9500. IEEE, 2023.
- Chin-Yew Lin. ROUGE: A package for automatic evaluation of summaries. In Text Summarization Branches Out, pp. 74–81, Barcelona, Spain, July 2004. Association for Computational Linguistics. URL https://aclanthology.org/W04-1013.
- Chameleon: Plug-and-play compositional reasoning with large language models, 2023.
- Fantastically ordered prompts and where to find them: Overcoming few-shot prompt order sensitivity. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 8086–8098, Dublin, Ireland, May 2022. Association for Computational Linguistics. doi: 10.18653/v1/2022.acl-long.556. URL https://aclanthology.org/2022.acl-long.556.
- Mathdial: A dialogue tutoring dataset with rich pedagogical properties grounded in math reasoning problems, 2023a.
- Opportunities and challenges in neural dialog tutoring. In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, pp. 2357–2372, Dubrovnik, Croatia, May 2023b. Association for Computational Linguistics. doi: 10.18653/v1/2023.eacl-main.173. URL https://aclanthology.org/2023.eacl-main.173.
- Using linguistic cues for the automatic recognition of personality in conversation and text. Journal of artificial intelligence research, 30:457–500, 2007.
- Webgpt: Browser-assisted question-answering with human feedback. CoRR, abs/2112.09332, 2021. URL https://arxiv.org/abs/2112.09332.
- OpenAI. Gpt-4 technical report, 2023.
- Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pp. 311–318, Philadelphia, Pennsylvania, USA, July 2002. Association for Computational Linguistics. doi: 10.3115/1073083.1073135. URL https://aclanthology.org/P02-1040.
- Matt Post. A call for clarity in reporting BLEU scores. In Proceedings of the Third Conference on Machine Translation: Research Papers, pp. 186–191, Brussels, Belgium, October 2018. Association for Computational Linguistics. doi: 10.18653/v1/W18-6319. URL https://aclanthology.org/W18-6319.
- Tool learning with foundation models, 2023a.
- Toolllm: Facilitating large language models to master 16000+ real-world apis. arXiv preprint arXiv:2307.16789, 2023b.
- The probabilistic relevance framework: Bm25 and beyond. Foundations and Trends® in Information Retrieval, 3(4):333–389, 2009.
- Hugginggpt: Solving ai tasks with chatgpt and its friends in huggingface. arXiv preprint arXiv:2303.17580, 2023.
- Cima: A large open access dialogue dataset for tutoring. In Proceedings of the Fifteenth Workshop on Innovative Use of NLP for Building Educational Applications, pp. 52–64, 2020.
- PsyQA: A Chinese dataset for generating long counseling text for mental health support. In Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pp. 1489–1503, Online, August 2021. Association for Computational Linguistics. doi: 10.18653/v1/2021.findings-acl.130. URL https://aclanthology.org/2021.findings-acl.130.
- Chain-of-thought prompting for responding to in-depth dialogue questions with llm, 2023a.
- Strategize before teaching: A conversational tutoring system with pedagogy self-distillation. In Findings of the Association for Computational Linguistics: EACL 2023, pp. 2268–2274, Dubrovnik, Croatia, May 2023b. Association for Computational Linguistics. URL https://aclanthology.org/2023.findings-eacl.170.
- Unleashing cognitive synergy in large language models: A task-solving agent through multi-persona self-collaboration. arXiv preprint arXiv:2307.05300, 2023c.
- Sherwood L Washburn. Tools and human evolution. Scientific American, 203(3):62–75, 1960.
- Chain-of-thought prompting elicits reasoning in large language models. In NeurIPS, 2022. URL http://papers.nips.cc/paper_files/paper/2022/hash/9d5609613524ecf4f15af0f7b31abca4-Abstract-Conference.html.
- A preliminary study of disputation behavior in online debating forum. In Proceedings of the Third Workshop on Argument Mining (ArgMining2016), pp. 166–171, Berlin, Germany, August 2016. Association for Computational Linguistics. doi: 10.18653/v1/W16-2820. URL https://aclanthology.org/W16-2820.
- Analyzing persuasion strategies of debaters on social media. In Proceedings of the 29th International Conference on Computational Linguistics, pp. 6897–6905, Gyeongju, Republic of Korea, October 2022. International Committee on Computational Linguistics. URL https://aclanthology.org/2022.coling-1.600.
- More is better: Enhancing open-domain dialogue generation via multi-source heterogeneous knowledge. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pp. 2286–2300, Online and Punta Cana, Dominican Republic, November 2021. Association for Computational Linguistics. doi: 10.18653/v1/2021.emnlp-main.175. URL https://aclanthology.org/2021.emnlp-main.175.
- Improving the applicability of knowledge-enhanced dialogue generation systems by using heterogeneous knowledge from multiple sources. In Proceedings of the Fifteenth ACM International Conference on WEB Search and Data Mining, pp. 1149–1157, 2022.
- Rewoo: Decoupling reasoning from observations for efficient augmented language models, 2023a.
- Long time no see! open-domain conversation with long-term persona memory. In Findings of the Association for Computational Linguistics: ACL 2022, pp. 2639–2650, Dublin, Ireland, May 2022. Association for Computational Linguistics. doi: 10.18653/v1/2022.findings-acl.207. URL https://aclanthology.org/2022.findings-acl.207.
- Exploring large language models for communication games: An empirical study on werewolf, 2023b.
- TAKE: Topic-shift aware knowledge sElection for dialogue generation. In Proceedings of the 29th International Conference on Computational Linguistics, pp. 253–265, Gyeongju, Republic of Korea, October 2022. International Committee on Computational Linguistics. URL https://aclanthology.org/2022.coling-1.20.
- React: Synergizing reasoning and acting in language models. In The Eleventh International Conference on Learning Representations, ICLR 2023, Kigali, Rwanda, May 1-5, 2023. OpenReview.net, 2023. URL https://openreview.net/pdf?id=WE_vluYUL-X.
- Few-shot conversational dense retrieval. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’21, pp. 829–838, New York, NY, USA, 2021. Association for Computing Machinery. ISBN 9781450380379. doi: 10.1145/3404835.3462856. URL https://doi.org/10.1145/3404835.3462856.
- Improving language models via plug-and-play retrieval feedback, 2023.
- Bertscore: Evaluating text generation with BERT. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net, 2020. URL https://openreview.net/forum?id=SkeHuCVFDr.
- Building emotional support chatbots in the era of llms. arXiv preprint arXiv:2308.11584, 2023.
- A dynamic strategy coach for effective negotiation. In Proceedings of the 20th Annual SIGdial Meeting on Discourse and Dialogue, pp. 367–378, Stockholm, Sweden, September 2019. Association for Computational Linguistics. doi: 10.18653/v1/W19-5943. URL https://aclanthology.org/W19-5943.
- Toolqa: A dataset for llm question answering with external tools, 2023.
- Hongru Wang (62 papers)
- Huimin Wang (24 papers)
- Lingzhi Wang (54 papers)
- Minda Hu (14 papers)
- Rui Wang (996 papers)
- Boyang Xue (23 papers)
- Hongyuan Lu (18 papers)
- Fei Mi (56 papers)
- Kam-Fai Wong (92 papers)