mlsanityData to 14 Oct 2026

Guiding Reasoning in Small Language Models with LLM Assistance

arXiv 2504.09923 ↗First posted 14 Apr 2025cs.CLcited by 14

On the map: LLM reasoning → Reasoning benchmarks and evaluation

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.

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.

Builds on

31 of 48 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025DeepSeek-R1 Thoughtology: Let's think about LLM Reasoning2 Apr

Other references · 30

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  2. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  3. 2023Let's Verify Step by Stepcited by 3,789
  4. 2023PaLM-E: An Embodied Multimodal Language Modelcited by 2,071
  5. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,986
  6. 2025s1: Simple test-time scalingcited by 1,341
  7. 2024Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMscited by 570
  8. 2024GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Modelscited by 550
  9. 2021DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Expertscited by 441
  10. 2024Deliberative Alignment: Reasoning Enables Safer Language Modelscited by 308
  11. 2024Diffusion Policy Policy Optimizationcited by 280
  12. 2024RLHF Workflow: From Reward Modeling to Online RLHFcited by 267
18 more
  1. 2024Chain-of-Thought Reasoning Without Promptingcited by 210
  2. 2024Don't Hallucinate, Abstain: Identifying LLM Knowledge Gaps via Multi-LLM Collaborationcited by 180
  3. 2023The Truth is in There: Improving Reasoning in Language Models with Layer-Selective Rank Reductioncited by 123
  4. 2024Diffusion-based Reinforcement Learning via Q-weighted Variational Policy Optimizationcited by 110
  5. 2024Arithmetic Without Algorithms: Language Models Solve Math With a Bag of Heuristicscited by 97
  6. 2024Small LLMs Are Weak Tool Learners: A Multi-LLM Agentcited by 93
  7. 2025Efficient Test-Time Scaling via Self-Calibrationcited by 75
  8. 2025ReasonFlux: Hierarchical LLM Reasoning via Scaling Thought Templatescited by 63
  9. 2024Learning to Decode Collaboratively with Multiple Language Modelscited by 62
  10. 2025CodeMonkeys: Scaling Test-Time Compute for Software Engineeringcited by 51
  11. 2025VersaPRM: Multi-Domain Process Reward Model via Synthetic Reasoning Datacited by 40
  12. 2025AdaptiveStep: Automatically Dividing Reasoning Step through Model Confidencecited by 24
  13. 2024Why do small language models underperform? Studying Language Model Saturation via the Softmax Bottleneckcited by 21
  14. 2025Minions: Cost-efficient Collaboration Between On-device and Cloud Language Modelscited by 20
  15. 2024Towards Fast Multilingual LLM Inference: Speculative Decoding and Specialized Drafterscited by 12
  16. 2025Evaluating Small Language Models for News Summarization: Implications and Factors Influencing Performancecited by 12
  17. 2024Enhancing the Reasoning Capabilities of Small Language Models via Solution Guidance Fine-Tuningcited by 11
  18. 2024SEER: Facilitating Structured Reasoning and Explanation via Reinforcement Learningcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 6; 2026: 8.

14 papers.

Sort:

By area:

  1. 2025Don't Overthink It: A Survey of Efficient R1-style Large Reasoning Modelscited by 37
  2. 2025MUR: Momentum Uncertainty guided Reasoning for Large Language Modelscited by 17
  3. 2026Confidence-Calibrated Small-Large Language Model Collaboration for Cost-Efficient Reasoningcited by 4
  4. 2026Select to Think: Unlocking SLM Potential with Local Sufficiencycited by 3
  5. 2025P/D-Device: Disaggregated Large Language Model between Cloud and Devicescited by 2
  6. 2025Investigating Language Model Capabilities to Represent and Process Formal Knowledge: A Preliminary Study to Assist Ontology Engineeringcited by 2
  7. 2025DICE: Structured Reasoning in LLMs through SLM-Guided Chain-of-Thought Correctioncited by 2
  8. 2025Mistake Notebook Learning: Batch-Clustered Failures for Training-Free Agent Adaptationcited by 2
  9. 2026SEF-CLGC at SemEval-2026 Task 11: Logical Notation Impact on Language Model Performancecited by 1
  10. 2026LegalDrill: Diagnosis-Driven Synthesis for Legal Reasoning in Small Language Modelscited by 0
  11. 2026Can Hallucinations Be Useful? Solving Multi-Hop Questions With SLMs By Chaining System-I/II Reasoningcited by 0
  12. 2026STRETCH the Boundaries: A Unified Self-Taught Framework for Progressive LLM Evolutioncited by 0
2 more
  1. 2026Offline Guidance, Online Reasoning: Reusing LLM Feedback for Small Language Modelscited by 0
  2. 2026Learning to Ask: Information Acquisition for SLM-LLM Collaboration, under a budgetcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026HintMR: Eliciting Stronger Mathematical Reasoning in Small Language Models
  2. 2025Towards Reasoning Ability of Small Language Models
  3. 2025Advancing Reasoning in Large Language Models: Promising Methods and Approaches
  4. 2025Enhancing Reasoning Capabilities of Small Language Models with Blueprints and Prompt Template Search
  5. 2026The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes
  6. 2024LLM Reasoners: New Evaluation, Library, and Analysis of Step-by-Step Reasoning with Large Language Models
  7. 2025Fast, Slow, and Tool-augmented Thinking for LLMs: A Review
  8. 2024Inference Scaling vs Reasoning: An Empirical Analysis of Compute-Optimal LLM Problem-Solving