Lune

EMNLP Findings2025

Exploring Deductive and Inductive Reasoning Capabilities of Large Language Models in Procedural Planning

Jiabao Kang, Xinye Li, Liyan Xu, Qingbin Liu, Xi Chen, Zhiying Tu, Dianhui Chu, Dianbo Sui

Abstract

Deductive and inductive reasoning are fundamental components of human cognition, and in daily life, people often apply these types of reasoning unconsciously. While previous studies have extensively examined the deductive and inductive reasoning abilities of Large Language Models (LLMs) in rule-based and math-related tasks, little attention has been given to their role in procedural planning--an area that holds considerable relevance for real-world applications. To fill this gap, we present DIRPP (Deductive and Inductive Reasoning in Procedural Planning) in this paper, a benchmark designed to assess the deductive and inductive reasoning abilities of various LLMs within the context of procedural planning. Based on the benchmark, we initially observe that LLMs demonstrate excellent deductive reasoning capabilities in procedural planning but show suboptimal performance in inductive reasoning. To enhance their inductive reasoning abilities, we further propose a novel and effective method called IMSE (Induction through Multiple Similar Examples), which enables LLMs to generate multiple similar procedural plans and then perform inductive reasoning based on these examples. Through various experiments, we find that the proposed method can significantly improve the inductive reasoning capabilities of LLMs. 1 /Task Description/ Given an abstract goal, generate two specific and concise goals related to it. Each goal should be as brief and straightforward as possible while adding relevant restrictions. Ensure that the specific goals differ from the example goal provided (i.e., Specific Goal) and focus on different aspects of the goal. /Examples/ Abstract goal: making a memory board Specific goal 1: making a memory board with notes Specific goal 2: make a memory board with photos Abstract goal: making photo blocks Specific goal 1: making photo blocks with friends Specific goal 2: making photo blocks with family Abstract goal: eating dragon fruit Specific goal 1: eating dragon fruit with a spoon Specific goal 2: eating dragon fruit topped with yogurt /Completion/ Abstract Goal: Abstract Goal Generated Specific Goals Table 18: An example of prompt for GPT-4o-mini for specific goals generation via in-context learning. Generated texts are highlighted. Specific Goal and Abstract Goal will be replaced with specific content.

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.

1 top-tier paper has cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Large Language Models as Commonsense Knowledge for Large-Scale Task Planning cites Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents
  • Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents cites Language Models are Few-Shot Learners
  • Leveraging Pre-trained Large Language Models to Construct and Utilize World Models for Model-based Task Planning cites Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents
  • Leveraging Pre-trained Large Language Models to Construct and Utilize World Models for Model-based Task Planning cites Language Models are Few-Shot Learners
  • Knowledge Vector of Logical Reasoning in Large Language Models cites Leveraging Pre-trained Large Language Models to Construct and Utilize World Models for Model-based Task Planning
  • Training language models to follow instructions with human feedback cites Language Models are Few-Shot Learners
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers1

  1. Knowledge Vector of Logical Reasoning in Large Language ModelsZixuan Wang, Yuanyuan LeiACL 2026

Builds on12

  1. Language Models are Few-Shot LearnersTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah et al.NeurIPS 2020 · 64,255 citations
  2. Training language models to follow instructions with human feedbackLong Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida et al.NeurIPS 2022 · 24,707 citations
  3. Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied AgentsWenlong Huang, Pieter Abbeel, Deepak Pathak, Igor MordatchICML 2022 · 1,539 citations
  4. Large Language Models as Commonsense Knowledge for Large-Scale Task PlanningZirui Zhao, Wee Sun Lee, David HsuNeurIPS 2023 · 423 citations
  5. Leveraging Pre-trained Large Language Models to Construct and Utilize World Models for Model-based Task PlanningLin Guan, Karthik Valmeekam, Sarath Sreedharan, Subbarao KambhampatiNeurIPS 2023 · 347 citations

Related papers

  1. InductionBench: LLMs Fail in the Simplest Complexity ClassWenyue Hua, Tyler Wong, Fei Sun, Liangming Pan et al.ACL 2025
  2. MME-Reasoning: A Broad-Spectrum Benchmark for Evaluating Logical Reasoning in MLLMsJiakang Yuan, Tianshuo Peng, Yilei Jiang, Yiting Lu et al.ICML 2026
  3. A Survey of Inductive Reasoning for Large Language ModelsKedi Chen, Dezhao Ruan, Yuhao Dan, Yaoting Wang et al.ACL 2026 · 5 citations
  4. Phenomenal Yet Puzzling: Testing Inductive Reasoning Capabilities of Language Models with Hypothesis RefinementLinlu Qiu, Liwei Jiang, Ximing Lu, Melanie Sclar et al.ICLR 2024 · 114 citations
  5. Hypothesis Search: Inductive Reasoning with Language ModelsRuocheng Wang, Eric Zelikman, Gabriel Poesia, Yewen Pu et al.ICLR 2024 · 156 citations
Dusk over the sea between two cliffs drawn in fine vertical lines