mlsanityData to 7 Oct 2026

Memory-assisted prompt editing to improve GPT-3 after deployment

arXiv 2201.06009 ↗First posted 16 Jan 2022cs.CLcited by 2

Not on the map. Why?

Abstract (arXiv)

Large LMs such as GPT-3 are powerful, but can commit mistakes that are obvious to humans. For example, GPT-3 would mistakenly interpret "What word is similar to good?" to mean a homophone, while the user intended a synonym. Our goal is to effectively correct such errors via user interactions with the system but without retraining, which will be prohibitively costly. We pair GPT-3 with a growing memory of recorded cases where the model misunderstood the user's intents, along with user feedback for clarification. Such a memory allows our system to produce enhanced prompts for any new query based on the user feedback for error correction on similar cases in the past. On four tasks (two lexical tasks, two advanced ethical reasoning tasks), we show how a (simulated) user can interactively teach a deployed GPT-3, substantially increasing its accuracy over the queries with different kinds of misunderstandings by the GPT-3. Our approach is a step towards the low-cost utility enhancement for very large pre-trained LMs. Code, data, and instructions to implement MEMPROMPT for a new task at https://www.memprompt.com/.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 26 of its 46 references that are in the map. First posted on arXiv 16 Jan 2022.

Builds on

26 of 46 references.

References · 26

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  3. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  4. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,049
  5. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,878
  6. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,747
  7. 2017Billion-scale similarity search with GPUscited by 2,477
  8. 2021Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processingcited by 2,222
  9. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,126
  10. 2021What Makes Good In-Context Examples for GPT-3?cited by 1,184
  11. 2019Generalization through Memorization: Nearest Neighbor Language Modelscited by 788
  12. 2021Editing Factual Knowledge in Language Modelscited by 548
14 more
  1. 2021Fast Model Editing at Scalecited by 491
  2. 2020Social Chemistry 101: Learning to Reason about Social and Moral Normscited by 253
  3. 2019Identifying and Reducing Gender Bias in Word-Level Language Modelscited by 214
  4. 2021How Many Data Points is a Prompt Worth?cited by 197
  5. 2020Generation-Augmented Retrieval for Open-domain Question Answeringcited by 186
  6. 2021Hallucinated but Factual! Inspecting the Factuality of Hallucinations in Abstractive Summarizationcited by 128
  7. 2021Do Language Models Have Beliefs? Methods for Detecting, Updating, and Visualizing Model Beliefscited by 79
  8. 2021BeliefBank: Adding Memory to a Pre-Trained Language Model for a Systematic Notion of Beliefcited by 56
  9. 2016Learning Language Games through Interactioncited by 43
  10. 2021Learning to Repair: Repairing model output errors after deployment using a dynamic memory of feedbackcited by 35
  11. 2021NL-EDIT: Correcting semantic parse errors through natural language interactioncited by 33
  12. 2021Evaluating Gender Bias in Natural Language Inferencecited by 27
  13. 2019Improving Question Answering with External Knowledgecited by 16
  14. 2016Integrating Distributional Lexical Contrast into Word Embeddings for Antonym-Synonym Distinctioncited by 13

Built on by

Cited by year

Citing papers per year: 2023: 1; 2024: 1.

2 papers.

  1. 2023Interactive Natural Language Processingcited by 31
  2. 2024Reasons and Solutions for the Decline in Model Performance after Editingcited by 7