mlsanityData to 7 Oct 2026

Complexity Agnostic Recursive Decomposition of Thoughts

arXiv 2601.04210 ↗First posted 10 Dec 2025cs.CL cs.AI cs.ITcited by 1

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

Large language models often fail on multi-step reasoning due to fixed reasoning strategies that ignore problem specific difficulty. We introduce CARD (Complexity Agnostic Recursive Decomposition), a framework that predicts problem complexity before generation and adapts decomposition accordingly. Our system comprises MRCE (Multi-dimensional Reasoning Complexity Estimator), a 0.6B Qwen model predicting 30 fine-grained features from question text and a two-stage recursive solver: (1) hierarchical decomposition into K steps based on task profile and (2) per-step thought budget allocation (1, 5-9, or 10 thoughts) via recursive MRCE profiling. Evaluated on three reasoning models (Qwen3-0.6B, DeepSeek-R1-Distill-Qwen-1.5B, Qwen3-1.7B), CARD achieves 81.4% to 89.2% accuracy on GSM8K while reducing token cost by 1.88x to 2.40x compared to fixed decomposition baselines. On MATH-500, CARD reaches 75.1 to 86.8% accuracy using 1.71x to 5.74x fewer tokens. Our results demonstrate that preemptive complexity estimation enables both higher accuracy and significant efficiency gains.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 13 of its 14 references that are in the map. First posted on arXiv 10 Dec 2025.

Builds on

13 of 14 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025Accelerating Training Speed of Tiny Recursive Models with Curriculum Guided Adaptive Recursion11 Nov
  2. 2024Token-Budget-Aware LLM Reasoning24 Dec
  3. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Models17 May

Other references · 10

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  3. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  4. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,203
  5. 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Modelscited by 1,397
  6. 2022Decomposed Prompting: A Modular Approach for Solving Complex Taskscited by 554
  7. 2023ADaPT: As-Needed Decomposition and Planning with Language Modelscited by 172
  8. 2024Estimating Knowledge in Large Language Models Without Generating a Single Tokencited by 39
  9. 2024On the Empirical Complexity of Reasoning and Planning in LLMscited by 6
  10. 2023ComplexityNet: Increasing LLM Inference Efficiency by Learning Task Complexitycited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026VERIFY-RL: Verifiable Recursive Decomposition for Reinforcement Learning in Mathematical Reasoningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Recursive Decomposition of Logical Thoughts: Framework for Superior Reasoning and Knowledge Propagation in Large Language Models
  2. 2025Encode, Think, Decode: Scaling test-time reasoning with recursive latent thoughts
  3. 2025THOUGHTTERMINATOR: Benchmarking, Calibrating, and Mitigating Overthinking in Reasoning Models
  4. 2023Recursion of Thought: A Divide-and-Conquer Approach to Multi-Context Reasoning with Language Models
  5. 2025Reasoning Effort and Problem Complexity: A Scaling Analysis in LLMs
  6. 2023Small Language Models Fine-tuned to Coordinate Larger Language Models improve Complex Reasoning
  7. 2026Playing Psychic: Using Thought Trees to Predict Reasoning Models Accuracy on Coding Tasks
  8. 2026How Much Thinking is Enough? Quantifying and Understanding Redundancy in LLM Reasoning