mlsanityData to 14 Oct 2026

Deciphering Trajectory-Aided LLM Reasoning: An Optimization Perspective

arXiv 2505.19815 ↗First posted 26 May 2025cs.CL cs.AIcited by 2

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

Abstract (arXiv)

We propose a novel framework for comprehending the reasoning capabilities of large language models (LLMs) through the perspective of meta-learning. By conceptualizing reasoning trajectories as pseudo-gradient descent updates to the LLM's parameters, we identify parallels between LLM reasoning and various meta-learning paradigms. We formalize the training process for reasoning tasks as a meta-learning setup, with each question treated as an individual task, and reasoning trajectories serving as the inner loop optimization for adapting model parameters. Once trained on a diverse set of questions, the LLM develops fundamental reasoning capabilities that can generalize to previously unseen questions. Extensive empirical evaluations substantiate the strong connection between LLM reasoning and meta-learning, exploring several issues of significant interest from a meta-learning standpoint. Our work not only enhances the understanding of LLM reasoning but also provides practical insights for improving these models through established meta-learning techniques.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 98 of its 127 references that are in the map. First posted on arXiv 26 May 2025.

Builds on

98 of 127 references.

Closest ancestors · 11

Same direction, newest first

  1. 2025Reasoning Models Can Be Effective Without Thinking14 Apr
  2. 2025Reasoning Models Know When They're Right: Probing Hidden States for Self-Verification7 Apr
  3. 2025Rethinking Reflection in Pre-Training5 Apr
  4. 2025A Survey of Efficient Reasoning for Large Reasoning Models: Language, Multimodality, and Beyond27 Mar
  5. 2025Towards Thinking-Optimal Scaling of Test-Time Compute for LLM Reasoning25 Feb
  6. 2025Atom of Thoughts for Markov LLM Test-Time Scaling17 Feb
  7. 2025s1: Simple test-time scaling31 Jan
  8. 2025A Survey on LLM Test-Time Compute via Search: Tasks, LLM Profiling, Search Algorithms, and Relevant Frameworks17 Jan
  9. 2024Imitate, Explore, and Self-Improve: A Reproduction Report on Slow-thinking Reasoning Systems12 Dec
  10. 2024O1 Replication Journey: A Strategic Progress Report -- Part 18 Oct
  11. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Models17 May

Other references · 87

  1. 2017Attention Is All You Needcited by 60,361
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  3. 2020Language Models are Few-Shot Learnerscited by 32,419
  4. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  5. 2023GPT-4 Technical Reportcited by 18,396
  6. 2022Training language models to follow instructions with human feedbackcited by 15,971
  7. 2017Proximal Policy Optimization Algorithmscited by 15,215
  8. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  9. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  10. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  11. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  12. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
38 more
  1. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  2. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  3. 2024Qwen2.5 Technical Reportcited by 4,942
  4. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  5. 2023Let's Verify Step by Stepcited by 3,789
  6. 2013Playing Atari with Deep Reinforcement Learningcited by 3,274
  7. 2017Prototypical Networks for Few-shot Learningcited by 3,015
  8. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,879
  9. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
  10. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 2,057
  11. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Codecited by 2,004
  12. 2022Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?cited by 1,434
  13. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,210
  14. 2024Qwen2.5-Coder Technical Reportcited by 1,203
  15. 2024DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligencecited by 1,181
  16. 2017Visualizing the Loss Landscape of Neural Netscited by 1,130
  17. 2024Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvementcited by 1,090
  18. 2022In-context Learning and Induction Headscited by 964
  19. 2019How Can We Know What Language Models Know?cited by 959
  20. 2020Meta-Learning in Neural Networks: A Surveycited by 950
  21. 2021An Explanation of In-context Learning as Implicit Bayesian Inferencecited by 938
  22. 2023Graph of Thoughts: Solving Elaborate Problems with Large Language Modelscited by 886
  23. 2022What learning algorithm is in-context learning? Investigations with linear modelscited by 684
  24. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 672
  25. 2021Learning To Retrieve Prompts for In-Context Learningcited by 648
  26. 2018Universal Transformerscited by 647
  27. 2016Learning to learn by gradient descent by gradient descentcited by 620
  28. 2019Meta-Learning with Implicit Gradientscited by 525
  29. 2025LIMO: Less is More for Reasoningcited by 498
  30. 2019Meta-Learning with Differentiable Convex Optimizationcited by 461
  31. 2025Open-Reasoner-Zero: An Open Source Approach to Scaling Up Reinforcement Learning on the Base Modelcited by 461
  32. 2021MetaICL: Learning to Learn In Contextcited by 454
  33. 2023Scaling Relationship on Learning Mathematical Reasoning with Large Language Modelscited by 427
  34. 2019Are Transformers universal approximators of sequence-to-sequence functions?cited by 406
  35. 2025Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Modelscited by 384
  36. 2014Memory Networkscited by 363
  37. 2023Iterative Preference Learning from Human Feedback: Bridging Theory and Practice for RLHF under KL-Constraintcited by 350
  38. 2023Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Modelscited by 349

50 most cited of 87 All 87 (JSON)

Built on by

Cited by year

Citing papers per year: 2025: 2.

2 papers.

  1. 2025Learning without training: The implicit dynamics of in-context learningcited by 43
  2. 2025Rectifying LLM Thought from Lens of Optimizationcited by 4

Similar, not cited

Close in wording, no citation either way

  1. 2024DOTS: Learning to Reason Dynamically in LLMs via Optimal Reasoning Trajectories Search
  2. 2026Exploring the Meta-level Reasoning of Large Language Models via a Tool-based Multi-hop Tabular Question Answering Task
  3. 2026The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes
  4. 2025Evaluating the Meta- and Object-Level Reasoning of Large Language Models for Question Answering
  5. 2025Language Models Coupled with Metacognition Can Outperform Reasoning Models
  6. 2026LLM Reasoning as Trajectories: Step-Specific Representation Geometry and Correctness Signals
  7. 2026Beyond Meta-Reasoning: Metacognitive Consolidation for Self-Improving LLM Reasoning
  8. 2025Fast, Slow, and Tool-augmented Thinking for LLMs: A Review