July 21, 2025Open Access

Toward Efficient and Faithful Reasoning in Large Language Models

View Full Paper

Ask AI

Bookmark

Authors

Discussion

This survey reveals methodologies to improve reasoning efficiency in LLMs, highlighting prompt-based methods and neuro-symbolic integration.

The survey categorizes approaches into prompt-based methods, architectural enhancements, and training-time techniques.
Key methodologies include chain-of-thought and retrieval-augmented generation for better reasoning accuracy.
It highlights the importance of interpretability in reasoning models while managing computational efficiency.
The paper emphasizes the need for future research on general-purpose reasoning agents and their open challenges.

Schneider et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark

This survey reveals methodologies to improve reasoning efficiency in LLMs, highlighting prompt-based methods and neuro-symbolic integration.