mlsanityData to 14 Oct 2026

Training-free LLM Verification via Recycling Few-shot Examples

arXiv 2506.17251 ↗First posted 8 Jun 2025cs.LG cs.AIcited by 4

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

Although large language models (LLMs) have achieved remarkable performance, the inherent stochasticity of their reasoning processes and varying conclusions present significant challenges. Majority voting or Best-of-N with external verifiers has been explored to mitigate this, but these approaches are limited in applicability or require additional training. To address this problem, we propose a novel framework that Recycles Few-shot examples to verify LLM outputs (ReFeri). Our key idea is to utilize the given few-shot examples not only to generate outputs, but also to evaluate the candidate outputs. Specifically, ReFeri combines a forward confidence score with a backward reconstruction penalty to select candidates that follow few-shot guidance while avoiding demonstration-specific overfitting. Experiments with three different LLMs across seven diverse tasks demonstrate that our framework significantly improves the accuracy of LLMs---achieving an average relative gain of 8.2%---through effective response selection.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 53 of its 68 references that are in the map. First posted on arXiv 8 Jun 2025.

Builds on

53 of 68 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025Scalable Best-of-N Selection for Large Language Models via Self-Certainty25 Feb
  2. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters6 Aug
  3. 2023Universal Self-Consistency for Large Language Model Generation29 Nov
  4. 2023Let's Sample Step by Step: Adaptive-Consistency for Efficient Reasoning and Coding with LLMs19 May
  5. 2022Language Models (Mostly) Know What They Know11 Jul

Other references · 48

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2024The Llama 3 Herd of Modelscited by 14,715
  3. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  4. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  5. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  6. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  7. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  8. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  9. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  10. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  11. 2022Large Language Models are Zero-Shot Reasonerscited by 4,374
  12. 2023Let's Verify Step by Stepcited by 3,789
36 more
  1. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  2. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,879
  3. 2020Learning to summarize from human feedbackcited by 2,588
  4. 2024Qwen2 Technical Reportcited by 2,120
  5. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Codecited by 2,004
  6. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,848
  7. 2022Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?cited by 1,434
  8. 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Modelscited by 1,407
  9. 2022Solving Quantitative Reasoning Problems with Language Modelscited by 1,392
  10. 2021What Makes Good In-Context Examples for GPT-3?cited by 1,195
  11. 2022Atlas: Few-shot Learning with Retrieval Augmented Language Modelscited by 952
  12. 2023Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotationscited by 928
  13. 2019DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphscited by 903
  14. 2022Solving math word problems with process- and outcome-based feedbackcited by 901
  15. 2024Self-Rewarding Language Modelscited by 681
  16. 2023Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Modelscited by 560
  17. 2024Generative Verifiers: Reward Modeling as Next-Token Predictioncited by 480
  18. 2023LaMP: When Large Language Models Meet Personalizationcited by 437
  19. 2022Complexity-Based Prompting for Multi-Step Reasoningcited by 388
  20. 2021True Few-Shot Learning with Language Modelscited by 344
  21. 2024To CoT or not to CoT? Chain-of-thought helps mainly on math and symbolic reasoningcited by 263
  22. 2023Better Zero-Shot Reasoning with Role-Play Promptingcited by 223
  23. 2024Chain-of-Thought Reasoning Without Promptingcited by 210
  24. 2023MuSR: Testing the Limits of Chain-of-thought with Multistep Soft Reasoningcited by 201
  25. 2024Recursive Introspection: Teaching Language Model Agents How to Self-Improvecited by 165
  26. 2022Self-Adaptive In-Context Learning: An Information Compression Perspective for In-Context Example Selection and Orderingcited by 161
  27. 2024The Impact of Reasoning Step Length on Large Language Modelscited by 153
  28. 2024Generative Reward Modelscited by 148
  29. 2024BoNBoN Alignment for Large Language Models and the Sweetness of Best-of-n Samplingcited by 121
  30. 2024Democratizing Large Language Models via Personalized Parameter-Efficient Fine-tuningcited by 90
  31. 2024SuRe: Summarizing Retrievals using Answer Candidates for Open-domain QA of LLMscited by 67
  32. 2024Revisiting Demonstration Selection Strategies in In-Context Learningcited by 58
  33. 2024In-Context Principle Learning from Mistakescited by 44
  34. 2024Few-shot Personalization of LLMs with Mis-aligned Responsescited by 32
  35. 2025Innate Reasoning is Not Enough: In-Context Learning Enhances Reasoning Large Language Models with Less Overthinkingcited by 20
  36. 2025Recitation over Reasoning: How Cutting-Edge Language Models Can Fail on Elementary School-Level Reasoning Problems?cited by 15

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 3.

4 papers.

  1. 2025Revisiting the UID Hypothesis in LLM Reasoning Tracescited by 0
  2. 2026A Training-Free Regeneration Paradigm: Contrastive Reflection Memory Guided Self-Verification and Self-Improvementcited by 0
  3. 2026LaRA: Layer-wise Representation Analysis for Detecting Data Contamination in RL Post-Trainingcited by 0
  4. 2026LookBack: Where and How to Score LVLM Responses via Visual Reference Usagecited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Less is More: Improving LLM Reasoning with Minimal Test-Time Intervention
  2. 2025Sample Smart, Not Hard: Correctness-First Decoding for Better Reasoning in LLMs
  3. 2025High Accuracy, Less Talk (HALT): Reliable LLMs through Capability-Aligned Finetuning
  4. 2024LLM Inference Unveiled: Survey and Roofline Model Insights
  5. 2025LLM Post-Training: A Deep Dive into Reasoning Large Language Models
  6. 2026PRISM: A Unified Framework for Post-Training LLMs Without Verifiable Rewards
  7. 2024The Surprising Effectiveness of Test-Time Training for Few-Shot Learning
  8. 2024Improving LLM Reasoning through Scaling Inference Computation with Collaborative Verification