# Guiding Reasoning in Small Language Models with LLM Assistance

arXiv: https://arxiv.org/abs/2504.09923 · HTML: https://mlsanity.com/paper/2504.09923/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 31 of its 48 references that are in the map.
- First posted on arXiv 14 Apr 2025.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 30 of 31)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [PaLM-E: An Embodied Multimodal Language Model](https://mlsanity.com/paper/2303.03378/) (2023)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [s1: Simple test-time scaling](https://mlsanity.com/paper/2501.19393/) (2025)
- [Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs](https://mlsanity.com/paper/2412.21187/) (2024)
- [GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models](https://mlsanity.com/paper/2410.05229/) (2024)
- [DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts](https://mlsanity.com/paper/2105.03023/) (2021)
- [Deliberative Alignment: Reasoning Enables Safer Language Models](https://mlsanity.com/paper/2412.16339/) (2024)
- [Diffusion Policy Policy Optimization](https://mlsanity.com/paper/2409.00588/) (2024)
- [RLHF Workflow: From Reward Modeling to Online RLHF](https://mlsanity.com/paper/2405.07863/) (2024)
- [Chain-of-Thought Reasoning Without Prompting](https://mlsanity.com/paper/2402.10200/) (2024)
- [Don't Hallucinate, Abstain: Identifying LLM Knowledge Gaps via Multi-LLM Collaboration](https://mlsanity.com/paper/2402.00367/) (2024)
- [The Truth is in There: Improving Reasoning in Language Models with Layer-Selective Rank Reduction](https://mlsanity.com/paper/2312.13558/) (2023)
- [Diffusion-based Reinforcement Learning via Q-weighted Variational Policy Optimization](https://mlsanity.com/paper/2405.16173/) (2024)
- [Arithmetic Without Algorithms: Language Models Solve Math With a Bag of Heuristics](https://mlsanity.com/paper/2410.21272/) (2024)
- [Small LLMs Are Weak Tool Learners: A Multi-LLM Agent](https://mlsanity.com/paper/2401.07324/) (2024)
- [DeepSeek-R1 Thoughtology: Let's think about LLM Reasoning](https://mlsanity.com/paper/2504.07128/) (2025)
- [Efficient Test-Time Scaling via Self-Calibration](https://mlsanity.com/paper/2503.00031/) (2025)
- [ReasonFlux: Hierarchical LLM Reasoning via Scaling Thought Templates](https://mlsanity.com/paper/2502.06772/) (2025)
- [Learning to Decode Collaboratively with Multiple Language Models](https://mlsanity.com/paper/2403.03870/) (2024)
- [CodeMonkeys: Scaling Test-Time Compute for Software Engineering](https://mlsanity.com/paper/2501.14723/) (2025)
- [VersaPRM: Multi-Domain Process Reward Model via Synthetic Reasoning Data](https://mlsanity.com/paper/2502.06737/) (2025)
- [AdaptiveStep: Automatically Dividing Reasoning Step through Model Confidence](https://mlsanity.com/paper/2502.13943/) (2025)
- [Why do small language models underperform? Studying Language Model Saturation via the Softmax Bottleneck](https://mlsanity.com/paper/2404.07647/) (2024)
- [Minions: Cost-efficient Collaboration Between On-device and Cloud Language Models](https://mlsanity.com/paper/2502.15964/) (2025)
- [Towards Fast Multilingual LLM Inference: Speculative Decoding and Specialized Drafters](https://mlsanity.com/paper/2406.16758/) (2024)
- [Evaluating Small Language Models for News Summarization: Implications and Factors Influencing Performance](https://mlsanity.com/paper/2502.00641/) (2025)
- [Enhancing the Reasoning Capabilities of Small Language Models via Solution Guidance Fine-Tuning](https://mlsanity.com/paper/2412.09906/) (2024)

## Built on by (most cited first, 14 of 14)

- [Don't Overthink It: A Survey of Efficient R1-style Large Reasoning Models](https://mlsanity.com/paper/2508.02120/) (2025)
- [MUR: Momentum Uncertainty guided Reasoning for Large Language Models](https://mlsanity.com/paper/2507.14958/) (2025)
- [Confidence-Calibrated Small-Large Language Model Collaboration for Cost-Efficient Reasoning](https://mlsanity.com/paper/2603.03752/) (2026)
- [Select to Think: Unlocking SLM Potential with Local Sufficiency](https://mlsanity.com/paper/2604.26940/) (2026)
- [P/D-Device: Disaggregated Large Language Model between Cloud and Devices](https://mlsanity.com/paper/2508.09035/) (2025)
- [Investigating Language Model Capabilities to Represent and Process Formal Knowledge: A Preliminary Study to Assist Ontology Engineering](https://mlsanity.com/paper/2509.10249/) (2025)
- [DICE: Structured Reasoning in LLMs through SLM-Guided Chain-of-Thought Correction](https://mlsanity.com/paper/2510.09211/) (2025)
- [Mistake Notebook Learning: Batch-Clustered Failures for Training-Free Agent Adaptation](https://mlsanity.com/paper/2512.11485/) (2025)
- [SEF-CLGC at SemEval-2026 Task 11: Logical Notation Impact on Language Model Performance](https://mlsanity.com/paper/2606.09157/) (2026)
- [LegalDrill: Diagnosis-Driven Synthesis for Legal Reasoning in Small Language Models](https://mlsanity.com/paper/2604.23809/) (2026)
- [Can Hallucinations Be Useful? Solving Multi-Hop Questions With SLMs By Chaining System-I/II Reasoning](https://mlsanity.com/paper/2605.27596/) (2026)
- [STRETCH the Boundaries: A Unified Self-Taught Framework for Progressive LLM Evolution](https://mlsanity.com/paper/2609.18642/) (2026)
- [Offline Guidance, Online Reasoning: Reusing LLM Feedback for Small Language Models](https://mlsanity.com/paper/2609.39346/) (2026)
- [Learning to Ask: Information Acquisition for SLM-LLM Collaboration, under a budget](https://mlsanity.com/paper/2610.01236/) (2026)

## Abstract (arXiv)

The limited reasoning capabilities of small language models (SLMs) cast doubt on their suitability for tasks demanding deep, multi-step logical deduction. This paper introduces a framework called Small Reasons, Large Hints (SMART), which selectively augments SLM reasoning with targeted guidance from large language models (LLMs). Inspired by the concept of cognitive scaffolding, SMART employs a score-based evaluation to identify uncertain reasoning steps and injects corrective LLM-generated reasoning only when necessary. By framing structured reasoning as an optimal policy search, our approach steers the reasoning trajectory toward correct solutions without exhaustive sampling. Our experiments on mathematical reasoning datasets demonstrate that targeted external scaffolding significantly improves performance, paving the way for collaborative use of both SLM and LLM to tackle complex reasoning tasks that are currently unsolvable by SLMs alone.