mlsanityData to 7 Oct 2026

Improving Probability-based Prompt Selection Through Unified Evaluation and Analysis

arXiv 2305.14877 ↗First posted 24 May 2023cs.CLcited by 13

Not on the map. Why?

Abstract (arXiv)

Previous works in prompt engineering for large language models have introduced different gradient-free probability-based prompt selection methods that aim to choose the optimal prompt among the candidates for a given task but have failed to provide a comprehensive and fair comparison between each other. In this paper, we propose a unified framework to interpret and evaluate the existing probability-based prompt selection methods by performing extensive experiments on 13 common and diverse NLP tasks. We find that each of the existing methods can be interpreted as some variant of the method that maximizes mutual information between the input and the predicted output (MI). Utilizing this finding, we develop several other combinatorial variants of MI and increase the effectiveness of the oracle prompt selection method from 87.79% to 94.98%, measured as the ratio of the performance of the selected prompt to that of the optimal oracle prompt. Furthermore, considering that all the methods rely on the output probability distribution of the model that might be biased, we propose a novel calibration method called Calibration by Marginalization (CBM) that is orthogonal to the existing methods and helps increase the prompt selection effectiveness of the best method to 96.85%, achieving 99.44% of the oracle prompt F1 without calibration.

Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). Builds on 38 of its 54 references that are in the map. First posted on arXiv 24 May 2023.

Builds on

38 of 54 references.

References · 38

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2023GPT-4 Technical Reportcited by 18,132
  3. 2018GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understandingcited by 4,854
  4. 2019HellaSwag: Can a Machine Really Finish Your Sentence?cited by 4,018
  5. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,878
  6. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,747
  7. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,464
  8. 2019PIQA: Reasoning about Physical Commonsense in Natural Languagecited by 2,818
  9. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,674
  10. 2015Character-level Convolutional Networks for Text Classificationcited by 2,293
  11. 2021Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processingcited by 2,222
  12. 2022BLOOM: A 176B-Parameter Open-Access Multilingual Language Modelcited by 1,968
26 more
  1. 2019SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systemscited by 1,718
  2. 2021Multitask Prompted Training Enables Zero-Shot Task Generalizationcited by 1,462
  3. 2021Calibrate Before Use: Improving Few-Shot Performance of Language Modelscited by 1,371
  4. 2020Making Pre-trained Language Models Better Few-shot Learnerscited by 1,255
  5. 2021What Makes Good In-Context Examples for GPT-3?cited by 1,184
  6. 2021Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivitycited by 1,122
  7. 2019How Can We Know What Language Models Know?cited by 953
  8. 2022A Survey on In-context Learningcited by 717
  9. 2021Learning To Retrieve Prompts for In-Context Learningcited by 641
  10. 2020TweetEval: Unified Benchmark and Comparative Evaluation for Tweet Classificationcited by 414
  11. 2022RLPrompt: Optimizing Discrete Text Prompts with Reinforcement Learningcited by 361
  12. 2022PromptSource: An Integrated Development Environment and Repository for Natural Language Promptscited by 288
  13. 2021Surface Form Competition: Why the Highest Probability Answer Isn't Always Rightcited by 256
  14. 2020AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Promptscited by 246
  15. 2022Demystifying Prompts in Language Models via Perplexity Estimationcited by 215
  16. 2022Active Example Selection for In-Context Learningcited by 195
  17. 2021Reframing Instructional Prompts to GPTk's Languagecited by 176
  18. 2022Self-Adaptive In-Context Learning: An Information Compression Perspective for In-Context Example Selection and Orderingcited by 158
  19. 2022GrIPS: Gradient-free, Edit-based Instruction Search for Prompting Large Language Modelscited by 127
  20. 2022Z-ICL: Zero-Shot In-Context Learning with Pseudo-Demonstrationscited by 70
  21. 2022Self-Generated In-Context Learning: Leveraging Auto-regressive Language Models as a Demonstration Generatorcited by 68
  22. 2022Data Curation Alone Can Stabilize In-context Learningcited by 58
  23. 2021Reordering Examples Helps during Priming-based Few-Shot Learningcited by 50
  24. 2022GPS: Genetic Prompt Search for Efficient Few-shot Learningcited by 37
  25. 2018Multi-Source Social Feedback of Online News Feedscited by 14
  26. 2022Zero-Label Prompt Selectioncited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 5; 2024: 5; 2025: 1; 2026: 2.

13 papers.

Sort:

By area:

  1. 2024InfoRM: Mitigating Reward Hacking in RLHF via Information-Theoretic Reward Modelingcited by 76
  2. 2024Smoothie: Label Free Language Model Routingcited by 31
  3. 2023Generative Calibration for In-context Learningcited by 20
  4. 2023InstructEval: Systematic Evaluation of Instruction Selection Methodscited by 12
  5. 2023Prompt Valuation Based on Shapley Valuescited by 10
  6. 2023Lever LM: Configuring In-Context Sequence to Lever Large Vision Language Modelscited by 6
  7. 2025Information-Theoretic Reward Modeling for Stable RLHF: Detecting and Mitigating Reward Hackingcited by 5
  8. 2024Exploiting the Vulnerability of Large Language Models via Defense-Aware Architectural Backdoorcited by 3
  9. 2024Elastic On-Device LLM Servicecited by 3
  10. 2023Unleashing the Multilingual Encoder Potential: Boosting Zero-Shot Performance via Probability Calibrationcited by 2
  11. 2024What Is Wrong with My Model? Identifying Systematic Problems with Semantic Data Slicingcited by 0
  12. 2026The Alignment Bottleneck in Decomposition-Based Claim Verificationcited by 0
1 more
  1. 2026Object Search in Partially-Known Environments via LLM-informed Model-based Planning and Prompt Selectioncited by 0