Thinking Machines: A Survey of LLM based Reasoning Strategies

2025-03-13

AI Summary

Weak reasoning capabilities of large language models (LLMs) and vision-language models (VLMs) hinder their trustworthy deployment in high-stakes domains such as healthcare, finance, and law. Method: We propose the first unified taxonomy for reasoning paradigms—systematically comparing chain-of-thought, tree-based, and reflective strategies—and analyze their fundamental trade-offs among generalizability, interpretability, and computational efficiency. Through a comprehensive survey of techniques—including chain-of-thought prompting, self-consistency, verifier-guided reasoning, process supervision, and VLM-augmented joint reasoning—we identify three core bottlenecks: long-horizon logical consistency, dynamic knowledge integration, and causal inference. Contribution/Results: We introduce novel evaluation dimensions tailored to trustworthy AI and establish a principled foundation for enhancing deep reasoning and real-world robustness in multimodal models, offering both theoretical insights and actionable methodological pathways.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Reasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Large language models for search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Large Language Models (LLMs) are highly proficient in language-based tasks. Their language capabilities have positioned them at the forefront of the future AGI (Artificial General Intelligence) race. However, on closer inspection, Valmeekam et al. (2024); Zecevic et al. (2023); Wu et al. (2024) highlight a significant gap between their language proficiency and reasoning abilities. Reasoning in LLMs and Vision Language Models (VLMs) aims to bridge this gap by enabling these models to think and re-evaluate their actions and responses. Reasoning is an essential capability for complex problem-solving and a necessary step toward establishing trust in Artificial Intelligence (AI). This will make AI suitable for deployment in sensitive domains, such as healthcare, banking, law, defense, security etc. In recent times, with the advent of powerful reasoning models like OpenAI O1 and DeepSeek R1, reasoning endowment has become a critical research topic in LLMs. In this paper, we provide a detailed overview and comparison of existing reasoning techniques and present a systematic survey of reasoning-imbued language models. We also study current challenges and present our findings.

Problem

• Bridging the gap between language proficiency and reasoning abilities in LLMs.
• Enhancing reasoning capabilities for complex problem-solving in AI.
• Surveying and comparing reasoning techniques in language models for sensitive domains.

Innovation

• Enhance LLMs reasoning for complex problem-solving.
• Survey reasoning techniques in language models.
• Address reasoning gaps in AI for sensitive domains.