ACL Findings2026
Thinking Economically: A Hierarchical Framework for Adaptive-Complexity Reasoning in LLMs
Yubo Gao, Haotian Wu, Hong Chen, Junquan Huang, Yibo Yan, Jungang Li, Zihao Dongfang, Sicheng Tao, Puay Siew Tan, Jie Zhang, Xuming Hu
Abstract
Chain-of-Thought (CoT) has significantly enhanced LLM reasoning, yet often incurs substantial computational overhead due to "overthinking": generating excessively long rationales without commensurate accuracy gains. Existing efficiency methods typically apply uniform compression, which overlooks a critical observation that reasoning complexity is heterogeneous at two distinct granularities: across different problems and within individual reasoning steps. This motivates our principle of Thinking Economically: intelligently allocating computational resources based on intrinsic task and step demands rather than pursuing uniform brevity. We propose Hierarchical Adaptive Budgeter (HAB), a training framework that operationalizes this principle through coarse-to-fine budgeting. At the interstep level, HAB predicts the optimal reasoning depth for each problem. At the intra-step level, HAB learns step-specific token budgeting signals from PPL-derived step comparisons and an adaptive Pareto optimization objective that captures the local quality-efficiency trade-off, while a Fisher Information-based pruner further provides fine-grained training-time guidance, thereby encouraging the generator to internalize more economical reasoning patterns. Experiments on GSM8K and MATH500 show that HAB not only surpasses standard CoT in accuracy but also reduces token usage, achieving a stronger performance-efficiency trade-off than the compared baselines.
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
- Tree of Thoughts: Deliberate Problem Solving with Large Language Models cites Self-Consistency Improves Chain of Thought Reasoning in Language Models
- Tree of Thoughts: Deliberate Problem Solving with Large Language Models cites Language Models are Few-Shot Learners
- STaR: Bootstrapping Reasoning With Reasoning cites Self-Consistency Improves Chain of Thought Reasoning in Language Models
- STaR: Bootstrapping Reasoning With Reasoning cites Language Models are Few-Shot Learners
- Self-Consistency Improves Chain of Thought Reasoning in Language Models cites Large Language Models are Zero-Shot Reasoners
- Large Language Models are Zero-Shot Reasoners cites Self-Consistency Improves Chain of Thought Reasoning in Language Models
- Self-Consistency Improves Chain of Thought Reasoning in Language Models cites Language Models are Few-Shot Learners
- Large Language Models are Zero-Shot Reasoners cites STaR: Bootstrapping Reasoning With Reasoning
- Large Language Models are Zero-Shot Reasoners cites Language Models are Few-Shot Learners
Builds on22
- Language Models are Few-Shot LearnersTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah et al.NeurIPS 2020 · 64,255 citations
- Large Language Models are Zero-Shot ReasonersTakeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo et al.NeurIPS 2022 · 8,168 citations
- Tree of Thoughts: Deliberate Problem Solving with Large Language ModelsShunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran et al.NeurIPS 2023 · 5,068 citations
- STaR: Bootstrapping Reasoning With ReasoningEric Zelikman, Yuhuai Wu, Jesse Mu, Noah D. GoodmanNeurIPS 2022 · 1,126 citations
- Self-Consistency Improves Chain of Thought Reasoning in Language ModelsXuezhi Wang, Jason Wei, Dale Schuurmans, Quoc V. Le et al.ICLR 2023 · 681 citations
Related papers
- Token-Budget-Aware LLM ReasoningTingxu Han, Zhenting Wang, Chunrong Fang, Shiyu Zhao et al.ACL Findings 2025 · 31 citations
- Plan and Budget: Effective and Efficient Test-Time Scaling on Reasoning Large Language ModelsJunhong Lin, Xinyue Zeng, Jie Zhu, Song Wang et al.ICLR 2026 · 30 citations
- RouteGoT: Node-Adaptive Routing for Cost-Efficient Graph of Thoughts ReasoningYuhang Liu, Ruijie Wang, Yunlong Chu, Bing Hao et al.KDD 2026 · 1 citation
- When More Thinking Hurts: Overthinking in LLM Test-Time Compute ScalingShu Zhou, Rui Ling, Junan Chen, Xin Wang et al.ACL Findings 2026
- From Long to Lean: Performance-aware and Adaptive Chain-of-Thought Compression via Multi-round RefinementJianzhi Yan, Le Liu, Youcheng Pan, Shiwei Chen et al.EMNLP 2025
