ACL Findings2025
Flowchart-Based Decision Making with Large Language Models
Abstract
Large language models (LLMs) are widely used for conversational systems, but they face significant challenges in interpretability of dialogue flow and reproducibility of expert knowledge. To address this, we propose a novel method that extracts flowcharts from dialogue data and incorporates them into LLMs. This approach not only makes the decision-making process more interpretable through visual representation, but also ensures the reproducibility of expert knowledge by explicitly modeling structured reasoning flows. By evaluating on dialogue datasets, we demonstrate that our method effectively reconstructs expert decisionmaking paths with high precision and recall scores. These findings underscore the potential of flowchart-based decision making to bridge the gap between flexibility and structured reasoning, making chatbot systems more interpretable for developers and end-users.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMs cites Direct Preference Optimization: Your Language Model is Secretly a Reward Model
- Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMs cites Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
- Bootstrapping LLM-based Task-Oriented Dialogue Agents via Self-Talk cites Direct Preference Optimization: Your Language Model is Secretly a Reward Model
Builds on7
- Retrieval-Augmented Generation for Knowledge-Intensive NLP TasksPatrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni et al.NeurIPS 2020 · 19,162 citations
- Direct Preference Optimization: Your Language Model is Secretly a Reward ModelRafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D. Manning et al.NeurIPS 2023 · 10,924 citations
- Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMsOded Ovadia, Menachem Brief, Moshik Mishaeli, Oren ElishaEMNLP 2024 · 89 citations
- Uncertainty of Thoughts: Uncertainty-Aware Planning Enhances Information Seeking in LLMsZhiyuan Hu, Chumin Liu, Xidong Feng, Yilun Zhao et al.NeurIPS 2024 · 43 citations
- Bootstrapping LLM-based Task-Oriented Dialogue Agents via Self-TalkDennis Ulmer, Elman Mansimov, Kaixiang Lin, Lijia Sun et al.ACL Findings 2024 · 13 citations
Related papers
- Unsupervised Extraction of Dialogue Policies from ConversationsMakesh Narsimhan Sreedhar, Traian Rebedea, Christopher ParisienEMNLP 2024 · 1 citation
- DecisionFlow: Advancing Large Language Model as Principled Decision MakerXiusi Chen, Shanyong Wang, Cheng Qian, Hongru Wang et al.EMNLP Findings 2025 · 5 citations
- Follow the Flow: Fine-grained Flowchart Attribution with Neurosymbolic AgentsManan Suri, Puneet Mathur, Nedim Lipka, Franck Dernoncourt et al.EMNLP 2025 · 1 citation
- METRO: Towards Strategy Induction from Expert Dialogue Transcripts for Non-collaborative DialoguesHaofu Yang, Jiaji Liu, Chen Huang, Faguo Wu et al.ACL 2026 · 2 citations
- Retro-Expert: Collaborative Reasoning for Interpretable RetrosynthesisXinyi Li, Sai Wang, Yutian Lin, Yu WuICML 2026 · 4 citations
