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

arXiv: https://arxiv.org/abs/2201.06009 · HTML: https://mlsanity.com/paper/2201.06009/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 46 references that are in the map.
- First posted on arXiv 16 Jan 2022.

## Builds on (most cited first, 26 of 26)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing](https://mlsanity.com/paper/2107.13586/) (2021)
- [REALM: Retrieval-Augmented Language Model Pre-Training](https://mlsanity.com/paper/2002.08909/) (2020)
- [What Makes Good In-Context Examples for GPT-3?](https://mlsanity.com/paper/2101.06804/) (2021)
- [Generalization through Memorization: Nearest Neighbor Language Models](https://mlsanity.com/paper/1911.00172/) (2019)
- [Editing Factual Knowledge in Language Models](https://mlsanity.com/paper/2104.08164/) (2021)
- [Fast Model Editing at Scale](https://mlsanity.com/paper/2110.11309/) (2021)
- [Social Chemistry 101: Learning to Reason about Social and Moral Norms](https://mlsanity.com/paper/2011.00620/) (2020)
- [Identifying and Reducing Gender Bias in Word-Level Language Models](https://mlsanity.com/paper/1904.03035/) (2019)
- [How Many Data Points is a Prompt Worth?](https://mlsanity.com/paper/2103.08493/) (2021)
- [Generation-Augmented Retrieval for Open-domain Question Answering](https://mlsanity.com/paper/2009.08553/) (2020)
- [Hallucinated but Factual! Inspecting the Factuality of Hallucinations in Abstractive Summarization](https://mlsanity.com/paper/2109.09784/) (2021)
- [Do Language Models Have Beliefs? Methods for Detecting, Updating, and Visualizing Model Beliefs](https://mlsanity.com/paper/2111.13654/) (2021)
- [BeliefBank: Adding Memory to a Pre-Trained Language Model for a Systematic Notion of Belief](https://mlsanity.com/paper/2109.14723/) (2021)
- [Learning Language Games through Interaction](https://mlsanity.com/paper/1606.02447/) (2016)
- [Learning to Repair: Repairing model output errors after deployment using a dynamic memory of feedback](https://mlsanity.com/paper/2112.09737/) (2021)
- [NL-EDIT: Correcting semantic parse errors through natural language interaction](https://mlsanity.com/paper/2103.14540/) (2021)
- [Evaluating Gender Bias in Natural Language Inference](https://mlsanity.com/paper/2105.05541/) (2021)
- [Improving Question Answering with External Knowledge](https://mlsanity.com/paper/1902.00993/) (2019)
- [Integrating Distributional Lexical Contrast into Word Embeddings for Antonym-Synonym Distinction](https://mlsanity.com/paper/1605.07766/) (2016)

## Built on by (most cited first, 2 of 2)

- [Interactive Natural Language Processing](https://mlsanity.com/paper/2305.13246/) (2023)
- [Reasons and Solutions for the Decline in Model Performance after Editing](https://mlsanity.com/paper/2410.23843/) (2024)

## 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/.