mlsanityData to 14 Oct 2026

Better Zero-Shot Reasoning with Self-Adaptive Prompting

arXiv 2305.14106 ↗First posted 23 May 2023cs.CL cs.AI cs.LGcited by 52

On the map: LLM reasoning → Prompting and chain-of-thought reasoning

Abstract (arXiv)

Modern large language models (LLMs) have demonstrated impressive capabilities at sophisticated tasks, often through step-by-step reasoning similar to humans. This is made possible by their strong few and zero-shot abilities -- they can effectively learn from a handful of handcrafted, completed responses ("in-context examples"), or are prompted to reason spontaneously through specially designed triggers. Nonetheless, some limitations have been observed. First, performance in the few-shot setting is sensitive to the choice of examples, whose design requires significant human effort. Moreover, given the diverse downstream tasks of LLMs, it may be difficult or laborious to handcraft per-task labels. Second, while the zero-shot setting does not require handcrafting, its performance is limited due to the lack of guidance to the LLMs. To address these limitations, we propose Consistency-based Self-adaptive Prompting (COSP), a novel prompt design method for LLMs. Requiring neither handcrafted responses nor ground-truth labels, COSP selects and builds the set of examples from the LLM zero-shot outputs via carefully designed criteria that combine consistency, diversity and repetition. In the zero-shot setting for three different LLMs, we show that using only LLM predictions, COSP improves performance up to 15% compared to zero-shot baselines and matches or exceeds few-shot baselines for a range of reasoning tasks.

Key facts. Cited by 52 papers in the map (data through 14 Oct 2026). Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning. Builds on 33 of its 50 references that are in the map. First posted on arXiv 23 May 2023.

Builds on

33 of 50 references.

Closest ancestors · 5

Same direction, newest first

  1. 2022Automatic Chain of Thought Prompting in Large Language Models7 Oct
  2. 2022Text and Patterns: For Effective Chain of Thought, It Takes Two to Tango16 Sep
  3. 2022Large Language Models are Zero-Shot Reasoners24 May
  4. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Models21 Mar
  5. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Models28 Jan

Other references · 28

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  3. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  4. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  5. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  6. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,326
  7. 2018GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understandingcited by 4,899
  8. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,724
  9. 2022Emergent Abilities of Large Language Modelscited by 2,512
  10. 2018CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledgecited by 1,822
  11. 2019SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systemscited by 1,729
  12. 2022LaMDA: Language Models for Dialog Applicationscited by 1,220
16 more
  1. 2021Scaling Language Models: Methods, Analysis & Insights from Training Gophercited by 1,202
  2. 2021What Makes Good In-Context Examples for GPT-3?cited by 1,195
  3. 2021Ethical and social risks of harm from Language Modelscited by 966
  4. 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 743
  5. 2022Large Language Models Can Self-Improvecited by 664
  6. 2021Learning To Retrieve Prompts for In-Context Learningcited by 648
  7. 2021Sentence-T5: Scalable Sentence Encoders from Pre-trained Text-to-Text Modelscited by 463
  8. 2022Unnatural Instructions: Tuning Language Models with (Almost) No Human Laborcited by 397
  9. 2022RLPrompt: Optimizing Discrete Text Prompts with Reinforcement Learningcited by 366
  10. 2017A Deep Reinforced Model for Abstractive Summarizationcited by 365
  11. 2016Solving General Arithmetic Word Problemscited by 360
  12. 2021In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learningcited by 301
  13. 2017Zero-Shot Learning -- The Good, the Bad and the Uglycited by 255
  14. 2017OpenNMT: Open-source Toolkit for Neural Machine Translationcited by 251
  15. 2022Selective Annotation Makes Language Models Better Few-Shot Learnerscited by 243
  16. 2016Learning values across many orders of magnitudecited by 85

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 13; 2024: 24; 2025: 10; 2026: 5.

52 papers.

Sort:

By area:

  1. 2023In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steeringcited by 245
  2. 2023Better Zero-Shot Reasoning with Role-Play Promptingcited by 223
  3. 2023Navigate through Enigmatic Labyrinth A Survey of Chain of Thought Reasoning: Advances, Frontiers and Futurecited by 184
  4. 2023Compositional Chain-of-Thought Prompting for Large Multimodal Modelscited by 164
  5. 2024Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluatorscited by 149
  6. 2023Batch Calibration: Rethinking Calibration for In-Context Learning and Prompt Engineeringcited by 94
  7. 2024In-context Learning with Retrieved Demonstrations for Language Models: A Surveycited by 70
  8. 2023Prompt Engineering a Prompt Engineercited by 53
  9. 2025Stepwise Perplexity-Guided Refinement for Efficient Chain-of-Thought Reasoning in Large Language Modelscited by 51
  10. 2024Multimodal Task Vectors Enable Many-Shot Multimodal In-Context Learningcited by 50
  11. 2024Efficient Prompting Methods for Large Language Models: A Surveycited by 47
  12. 2024Instance-adaptive Zero-shot Chain-of-Thought Promptingcited by 26
38 more
  1. 2024Fairer Preferences Elicit Improved Human-Aligned Large Language Model Judgmentscited by 25
  2. 2024MC-CoT: A Modular Collaborative CoT Framework for Zero-shot Medical-VQA with LLM and MLLM Integrationcited by 21
  3. 2025From Few to Many: Self-Improving Many-Shot Reasoners Through Iterative Optimization and Generationcited by 21
  4. 2025Towards LLMs Robustness to Changes in Prompt Format Stylescited by 21
  5. 2023Survival of the Most Influential Prompts: Efficient Black-Box Prompt Search via Clustering and Pruningcited by 17
  6. 2024Can We Verify Step by Step for Incorrect Answer Detection?cited by 15
  7. 2023Universal Self-Adaptive Promptingcited by 14
  8. 2024VisionGPT: LLM-Assisted Real-Time Anomaly Detection for Safe Visual Navigationcited by 14
  9. 2024Distilling Reasoning Ability from Large Language Models with Adaptive Thinkingcited by 13
  10. 2023Generalizable Chain-of-Thought Prompting in Mixed-task Scenarios with Large Language Modelscited by 11
  11. 2024Unleashing the Power of Large Language Models in Zero-shot Relation Extraction via Self-Promptingcited by 10
  12. 2023Towards Informative Few-Shot Prompt with Maximum Information Gain for In-Context Learningcited by 9
  13. 2024Self-Evolving GPT: A Lifelong Autonomous Experiential Learnercited by 9
  14. 2025What Really Matters in Many-Shot Attacks? An Empirical Study of Long-Context Vulnerabilities in LLMscited by 8
  15. 2023Test-Time Self-Adaptive Small Language Models for Question Answeringcited by 7
  16. 2024Learning from Contrastive Prompts: Automated Optimization and Adaptationcited by 7
  17. 2024Semi-Supervised Dialogue Abstractive Summarization via High-Quality Pseudolabel Selectioncited by 5
  18. 2024Self-Instructed Derived Prompt Generation Meets In-Context Learning: Unlocking New Potential of Black-Box LLMscited by 5
  19. 2025Thunder-NUBench: A Benchmark for LLMs' Sentence-Level Negation Understandingcited by 5
  20. 2023Large Language Models are In-context Teachers for Knowledge Reasoningcited by 4
  21. 2023The Tyranny of Possibilities in the Design of Task-Oriented LLM Systems: A Scoping Surveycited by 2
  22. 2024Grammatical Error Correction for Low-Resource Languages: The Case of Zarmacited by 2
  23. 2024PICLe: Pseudo-Annotations for In-Context Learning in Low-Resource Named Entity Detectioncited by 2
  24. 2026EMS: Multi-Agent Voting via Efficient Majority-then-Stoppingcited by 2
  25. 2026HCP-MAD:Heterogeneous Consensus-Progressive Reasoning for Efficient Multi-Agent Debatecited by 2
  26. 2026PromptAudit: Auditing Prompt Sensitivity in LLM-Based Vulnerability Detectioncited by 2
  27. 2024Zero-Shot Reasoning: Personalized Content Generation Without the Cold Start Problemcited by 1
  28. 2024SHROOM-INDElab at SemEval-2024 Task 6: Zero- and Few-Shot LLM-Based Classification for Hallucination Detectioncited by 1
  29. 2024Llama-VITS: Enhancing TTS Synthesis with Semantic Awarenesscited by 1
  30. 2024SSP: Self-Supervised Prompting for Cross-Lingual Transfer to Low-Resource Languages using Large Language Modelscited by 1
  31. 2024In-context Demonstration Matters: On Prompt Optimization for Pseudo-Supervision Refinementcited by 1
  32. 2024LLaMa-SciQ: An Educational Chatbot for Answering Science MCQcited by 0
  33. 2024STAYKATE: Hybrid In-Context Example Selection Combining Representativeness Sampling and Retrieval-based Approach -- A Case Study on Science Domainscited by 0
  34. 2025Efficient Knowledge Feeding to Language Models: A Novel Integrated Encoder-Decoder Architecturecited by 0
  35. 2025Evaluation of the Automated Labeling Method for Taxonomic Nomenclature Through Prompt-Optimized Large Language Modelcited by 0
  36. 2025Name of Thrones: Evaluating How LLMs Rank Student Names, Race, and Gender in Status Hierarchiescited by 0
  37. 2025LLMs as Architects and Critics for Multi-Source Opinion Summarizationcited by 0
  38. 2025GEM: Generative Entropy-Guided Preference Modeling for Few-shot Alignment of LLMscited by 0

50 most cited of 52 All 52 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Not the Example, but the Process: How Self-Generated Examples Enhance LLM Reasoning
  2. 2024Leveraging Zero-Shot Prompting for Efficient Language Model Distillation
  3. 2023In-Context Learning with Iterative Demonstration Selection
  4. 2023Self-ICL: Zero-Shot In-Context Learning with Self-Generated Demonstrations
  5. 2025Zero-Shot Verification-guided Chain of Thoughts
  6. 2025Reasoning Capabilities and Invariability of Large Language Models
  7. 2025Zero-Shot Commonsense Validation and Reasoning with Large Language Models: An Evaluation on SemEval-2020 Task 4 Dataset
  8. 2024Zero-Shot Chain-of-Thought Reasoning Guided by Evolutionary Algorithms in Large Language Models