ACL Findings2026
Time-for-Accuracy: Formalizing Chain-of-Thought as an Expansion of Logical Depth
Yue Wang, Zhi Zhang, Wang Xi, Chengjie Sun, Lili Shan, Bingquan Liu
Abstract
Chain-of-thought (CoT) often improves multistep reasoning, but it remains unclear what kind of additional sequential computation longer traces actually enable. We connect CoT to Bennett's logical depth, separating an answer's description length from the sequential effort required to derive it, and view a CoT budget of T steps as a qualitative cap on realizable sequential computation. To operationalize realized depth beyond raw length, we introduce Effective Logical Depth (ELD), a deletion-based measure of step necessity under a specified inference interface. Across depth-controlled prefix-sum tasks and GSM8K rationale perturbations, we observe two consistent signatures of a Time-for-Accuracy tradeoff: (i) plateauto-transition accuracy curves as the budget increases from being below to matching the task's required depth, and (ii) sparse, positiondependent deletion sensitivity concentrated in early steps for deeper instances. On GSM8K, an EXTRACT interface, where the model reads off the answer from the remaining rationale, remains near-perfect even after prefix deletions, whereas a REPAIR interface, where the model must re-solve from truncated rationale context, degrades markedly. Moreover, SOCRATIC human rationales are consistently more robust than MAIN rationales under REPAIR. These results suggest that longer CoT helps primarily when it enables additional effective sequential computation, and that deletion-based diagnostics can distinguish computational steps from redundant ones.
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
- 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
- Chain of Thoughtlessness? An Analysis of CoT in Planning cites Self-Consistency Improves Chain of Thought Reasoning in Language Models
- Chain of Thoughtlessness? An Analysis of CoT in Planning cites Large Language Models are Zero-Shot Reasoners
- AlphaMath Almost Zero: Process Supervision without Process cites Self-Consistency Improves Chain of Thought Reasoning in Language Models
Builds on18
- Large Language Models are Zero-Shot ReasonersTakeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo et al.NeurIPS 2022 · 8,168 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
- AlphaMath Almost Zero: Process Supervision without ProcessGuoxin Chen, Minpeng Liao, Chengxi Li, Kai FanNeurIPS 2024 · 219 citations
- ProcessBench: Identifying Process Errors in Mathematical ReasoningChujie Zheng, Zhenru Zhang, Beichen Zhang, Runji Lin et al.ACL 2025 · 209 citations
- Chain of Thoughtlessness? An Analysis of CoT in PlanningKaya Stechly, Karthik Valmeekam, Subbarao KambhampatiNeurIPS 2024 · 156 citations
Related papers
- Effective Reasoning Chains Reduce Intrinsic DimensionalityArchiki Prasad, Mandar Joshi, Kenton Lee, Mohit Bansal et al.ICML 2026 · 3 citations
- Thinking Economically: A Hierarchical Framework for Adaptive-Complexity Reasoning in LLMsYubo Gao, Haotian Wu, Hong Chen, Junquan Huang et al.ACL Findings 2026
- System-1.5 Reasoning: Traversal in Language and Latent Spaces with Dynamic ShortcutsXiaoqiang Wang, Suyuchen Wang, Yun Zhu, Bang LiuNeurIPS 2025 · 26 citations
- SyncThink: A Training-Free Strategy to Align Inference Termination with Reasoning SaturationGengyang Li, Wang Cai, Yifeng Gao, Yunfang WuACL Findings 2026
- The Impact of Reasoning Step Length on Large Language ModelsMingyu Jin, Qinkai Yu, Dong Shu, Haiyan Zhao et al.ACL Findings 2024 · 39 citations
