A Survey on LLM Inference-Time Self-Improvement
It organizes existing research for researchers in natural language processing, but is incremental as a survey without new results.
The paper surveys LLM inference-time self-improvement techniques, categorizing them into independent, context-aware, and model-aided approaches, and provides a review, taxonomy, and discussion of challenges.
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.