mlsanityData to 7 Oct 2026

A Survey on LLM Inference-Time Self-Improvement

arXiv 2412.14352 ↗First posted 18 Dec 2024cs.CLcited by 17

On the map: Synthetic data and self-improvement for LLMs → Self-rewarding alignment and self-training

Abstract (arXiv)

Techniques that enhance inference through increased computation at test-time have recently gained attention. In this survey, we investigate the current state of LLM Inference-Time Self-Improvement from three different perspectives: Independent Self-improvement, focusing on enhancements via decoding or sampling methods; Context-Aware Self-Improvement, leveraging additional context or datastore; and Model-Aided Self-Improvement, achieving improvement through model collaboration. We provide a comprehensive review of recent relevant studies, contribute an in-depth taxonomy, and discuss challenges and limitations, offering insights for future research.

Key facts. Cited by 17 papers in the map (data through 7 Oct 2026). Placed in Self-rewarding alignment and self-training, one of 13 directions in Synthetic data and self-improvement for LLMs. Builds on 13 of its 124 references that are in the map. First posted on arXiv 18 Dec 2024.

Builds on

13 of 124 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Mind the Gap: Examining the Self-Improvement Capabilities of Large Language Models3 Dec
  2. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters6 Aug

Other references · 11

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2020Scaling Laws for Neural Language Modelscited by 6,411
  3. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  4. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 2,950
  5. 2024Large Language Monkeys: Scaling Inference Compute with Repeated Samplingcited by 946
  6. 2023Gemini: A Family of Highly Capable Multimodal Modelscited by 810
  7. 2017Breaking the Softmax Bottleneck: A High-Rank RNN Language Modelcited by 147
  8. 2024Rethinking Interpretability in the Era of Large Language Modelscited by 110
  9. 2024Disclosure and Mitigation of Gender Bias in LLMscited by 47
  10. 2024ReasoningRec: Bridging Personalized Recommendations and Human-Interpretable Explanations through LLM Reasoningcited by 14
  11. 2023The Stable Entropy Hypothesis and Entropy-Aware Decoding: An Analysis and Algorithm for Robust Natural Language Generationcited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 14; 2026: 3.

17 papers.

Sort:

By area:

  1. 2025LLM Post-Training: A Deep Dive into Reasoning Large Language Modelscited by 129
  2. 2025A Survey of Context Engineering for Large Language Modelscited by 120
  3. 2025Reasoning Language Models: A Blueprintcited by 41
  4. 2025ConfTuner: Training Large Language Models to Express Their Confidence Verballycited by 37
  5. 2025Agents of Change: Self-Evolving LLM Agents for Strategic Planningcited by 20
  6. 2025Will Pre-Training Ever End? A First Step Toward Next-Generation Foundation MLLMs via Self-Improving Systematic Cognitioncited by 7
  7. 2025Self-Improvement in Multimodal Large Language Models: A Surveycited by 6
  8. 2025Neural Genetic Search in Discrete Spacescited by 5
  9. 2026Self-Improvement of Large Language Models: A Technical Overview and Future Outlookcited by 5
  10. 2025TestNUC: Enhancing Test-Time Computing Approaches and Scaling through Neighboring Unlabeled Data Consistencycited by 4
  11. 2025A Solver-in-the-Loop Framework for Improving LLMs on Answer Set Programming for Logic Puzzle Solvingcited by 4
  12. 2025ProtTeX: Structure-In-Context Reasoning and Editing of Proteins with Large Language Modelscited by 3
5 more
  1. 2025SDA: Steering-Driven Distribution Alignment for Open LLMs without Fine-Tuningcited by 2
  2. 2026When Languages Disagree: Self-Evolving Multilingual LLM Judgescited by 2
  3. 2025Towards a Small Language Model Lifecycle Frameworkcited by 0
  4. 2025Textual Self-attention Network: Test-Time Preference Optimization through Textual Gradient-based Attentioncited by 0
  5. 2026The Role of Disfluencies in Speech Translationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Refining Over Resampling: Test-Time Self-Correction for LLM Reasoning
  2. 2025Let's (not) just put things in Context: Test-Time Training for Long-Context LLMs
  3. 2024Faster and Lighter LLMs: A Survey on Current Challenges and Way Forward
  4. 2025Less is More: Improving LLM Reasoning with Minimal Test-Time Intervention
  5. 2026A Survey on Self-Improving Test-Time Intelligence: Feedback-Driven Adapting, Learning, and Scaling at Inference
  6. 2025DISC: Dynamic Decomposition Improves LLM Inference Scaling
  7. 2026A rigor-matched audit of periodic-step layer skipping for efficient llm inference: conflayers versus swift, with a supplemental analysis of trained routing alternatives
  8. 2025RoBoN: Routed Online Best-of-n for Test-Time Scaling with Multiple LLMs