Prompt engineering

Prompt engineering is the process of structuring or crafting an instruction in order to produce the best possible output from a generative artificial intelligence (AI) model.^[1]

Not to be confused with Command prompt.

A prompt is natural language text describing the task that an AI should perform.^[2] A prompt for a text-to-text language model can be a query, a command, or a longer statement including context, instructions, and conversation history. Prompt engineering may involve phrasing a query, specifying a style, choice of words and grammar,^[3] providing relevant context, or describing a character for the AI to mimic.^[1]

When communicating with a text-to-image or a text-to-audio model, a typical prompt is a description of a desired output such as "a high-quality photo of an astronaut riding a horse"^[4] or "Lo-fi slow BPM electro chill with organic samples".^[5] Prompting a text-to-image model may involve adding, removing, emphasizing, and re-ordering words to achieve a desired subject, style,^[6] layout, lighting,^[7] and aesthetic.

History

In 2018, researchers first proposed that all previously separate tasks in natural language processing (NLP) could be cast as a question-answering problem over a context. In addition, they trained a first single, joint, multi-task model that would answer any task-related question like "What is the sentiment" or "Translate this sentence to German" or "Who is the president?"^[8]

The AI boom saw an increase in the amount of "prompting technique" to get the model to output the desired outcome and avoid nonsensical output, a process characterized by trial-and-error.^[9] After the release of ChatGPT in 2022, prompt engineering was soon seen as an important business skill, albeit one with an uncertain economic future.^[1]

A repository for prompts reported that over 2,000 public prompts for around 170 datasets were available in February 2022.^[10] In 2022, the chain-of-thought prompting technique was proposed by Google researchers.^[11][12] In 2023, several text-to-text and text-to-image prompt databases were made publicly available.^[13][14] The Personalized Image-Prompt (PIP) dataset, a generated image-text dataset that has been categorized by 3,115 users, has also been made available publicly in 2024.^[15]

Text-to-text

Multiple distinct prompt engineering techniques have been published.

Chain-of-thought

According to Google Research, chain-of-thought (CoT) prompting is a technique that allows large language models (LLMs) to solve a problem as a series of intermediate steps before giving a final answer. In 2022, the Brain team of Google also claimed that chain-of-thought prompting improves reasoning ability by inducing the model to answer a multi-step problem with steps of reasoning that mimic a train of thought.^[11][16] Chain-of-thought techniques hypothetically allow large language models to overcome difficulties with some reasoning tasks that require logical thinking and multiple steps to solve, such as arithmetic or commonsense reasoning questions, according to announcements from Google and Amazon.^[17][18]

For example, given the question, "Q: The cafeteria had 23 apples. If they used 20 to make lunch and bought 6 more, how many apples do they have?", Google claims that a CoT prompt might induce the LLM to answer "A: The cafeteria had 23 apples originally. They used 20 to make lunch. So they had 23 - 20 = 3. They bought 6 more apples, so they have 3 + 6 = 9. The answer is 9."^[11] When applied to PaLM, a 540 billion parameter language model, Google claims that CoT prompting significantly aided the model, allowing it to perform comparably with task-specific fine-tuned models on several tasks, achieving state-of-the-art results at the time on the GSM8K mathematical reasoning benchmark.^[11] According to Google, it is possible to fine-tune models on CoT reasoning datasets to enhance this capability further and stimulate better interpretability.^[19][20]

An example of a CoT prompting:^[21]

   Q: {question}
   A: Let's think step by step.

As originally proposed by Google,^[11] each CoT prompt included a few Q&A examples. This made it a few-shot prompting technique. However, according to researchers at Google and the University of Tokyo, simply appending the words "Let's think step-by-step",^[21] has also proven effective, which makes CoT a zero-shot prompting technique. OpenAI claims that this prompt allows for better scaling as a user no longer needs to formulate many specific CoT Q&A examples.^[22]

In-context learning

In-context learning, refers to a model's ability to temporarily learn from prompts. For example, a prompt may include a few examples for a model to learn from, such as asking the model to complete "maison → house, chat → cat, chien →" (the expected response being dog),^[23] an approach called few-shot learning.^[24]

In-context learning is an emergent ability^[25] of large language models. It is an emergent property of model scale, meaning that breaks^[26] in downstream scaling laws occur, leading to its efficacy increasing at a different rate in larger models than in smaller models.^[27][11] Unlike training and fine-tuning, which produce lasting changes, in-context learning is temporary.^[28] Training models to perform in-context learning can be viewed as a form of meta-learning, or "learning to learn".^[29]

Self-consistency decoding

Self-consistency decoding^[30] performs several chain-of-thought rollouts, then selects the most commonly reached conclusion out of all the rollouts. If the rollouts disagree by a lot, a human can be queried for the correct chain of thought.^[31]

Tree-of-thought

Tree-of-thought prompting generalizes chain-of-thought by prompting the model to generate one or more "possible next steps", and then running the model on each of the possible next steps by breadth-first, beam, or some other method of tree search.^[32] The LLM has additional modules that can converse the history of the problem-solving process to the LLM, which allows the system to 'backtrack steps' the problem-solving process.

Prompting to disclose uncertainty

By default, the output of language models may not contain estimates of uncertainty. The model may output text that appears confident, though the underlying token predictions have low likelihood scores. Large language models like GPT-4 can have accurately calibrated likelihood scores in their token predictions,^[33] and so the model output uncertainty can be directly estimated by reading out the token prediction likelihood scores.

Prompting to estimate model sensitivity

Research consistently demonstrates that LLMs are highly sensitive to subtle variations in prompt formatting, structure, and linguistic properties. Some studies have shown up to 76 accuracy points across formatting changes in few-shot settings.^[34] Linguistic features significantly influence prompt effectiveness—such as morphology, syntax, and lexico-semantic changes—which meaningfully enhance task performance across a variety of tasks.^[3][35] Clausal syntax, for example, improves consistency and reduces uncertainty in knowledge retrieval.^[36] This sensitivity persists even with larger model sizes, additional few-shot examples, or instruction tuning.

To address sensitivity of models and make them more robust, several methods have been proposed. FormatSpread facilitates systematic analysis by evaluating a range of plausible prompt formats, offering a more comprehensive performance interval.^[34] Similarly, PromptEval estimates performance distributions across diverse prompts, enabling robust metrics such as performance quantiles and accurate evaluations under constrained budgets.^[37]

Automatic prompt generation

Retrieval-augmented generation

Main article: Retrieval-augmented generation

Two-phase process of document retrieval using dense embeddings and LLM for answer formulation

Retrieval-augmented generation (RAG) is a two-phase process involving document retrieval and answer generation by a large language model. The initial phase uses dense embeddings to retrieve documents. This retrieval can be based on a variety of database formats depending on the use case, such as a vector database, summary index, tree index, or keyword table index.^[38] In response to a query, a document retriever selects the most relevant documents. This relevance is typically determined by first encoding both the query and the documents into vectors, then identifying documents whose vectors are closest in Euclidean distance to the query vector. Following document retrieval, the LLM generates an output that incorporates information from both the query and the retrieved documents.^[39] RAG can also be used as a few-shot learner.

Graph retrieval-augmented generation

GraphRAG with a knowledge graph combining access patterns for unstructured, structured, and mixed data

GraphRAG^[40] (coined by Microsoft Research) is a technique that extends RAG with the use of a knowledge graph (usually, LLM-generated) to allow the model to connect disparate pieces of information, synthesize insights, and holistically understand summarized semantic concepts over large data collections. It was shown to be effective on datasets like the Violent Incident Information from News Articles (VIINA).^[41]

Earlier work showed the effectiveness of using a knowledge graph for question answering using text-to-query generation.^[42] These techniques can be combined to search across both unstructured and structured data, providing expanded context, and improved ranking.

Using language models to generate prompts

Large language models (LLM) themselves can be used to compose prompts for large language models.^[43] The automatic prompt engineer algorithm uses one LLM to beam search over prompts for another LLM:^[44][45]

• There are two LLMs. One is the target LLM, and another is the prompting LLM.
• Prompting LLM is presented with example input-output pairs, and asked to generate instructions that could have caused a model following the instructions to generate the outputs, given the inputs.
• Each of the generated instructions is used to prompt the target LLM, followed by each of the inputs. The log-probabilities of the outputs are computed and added. This is the score of the instruction.
• The highest-scored instructions are given to the prompting LLM for further variations.
• Repeat until some stopping criteria is reached, then output the highest-scored instructions.

CoT examples can be generated by LLM themselves. In "auto-CoT",^[46] a library of questions are converted to vectors by a model such as BERT. The question vectors are clustered. Questions nearest to the centroids of each cluster are selected. An LLM does zero-shot CoT on each question. The resulting CoT examples are added to the dataset. When prompted with a new question, CoT examples to the nearest questions can be retrieved and added to the prompt.

Text-to-image

See also: Artificial intelligence art § Prompt engineering and sharing

Example of prompt engineering for text-to-image generation, with Fooocus

In 2022, text-to-image models like DALL-E 2, Stable Diffusion, and Midjourney were released to the public.^[47] These models take text prompts as input and use them to generate AI-generated images. Text-to-image models typically do not understand grammar and sentence structure in the same way as large language models,^[48] thus may require a different set of prompting techniques.

Text-to-image models do not natively understand negation. The prompt "a party with no cake" is likely to produce an image including a cake.^[48] As an alternative, negative prompts allow a user to indicate, in a separate prompt, which terms should not appear in the resulting image.^[49] Techniques such as framing the normal prompt into a sequence-to-sequence language modeling problem can be used to automatically generate an output for the negative prompt.^[50]

Demonstration of the effect of negative prompts on images generated with Stable Diffusion

• Top: no negative prompt
• Centre: "green trees"
• Bottom: "round stones, round rocks"

Prompt formats

A text-to-image prompt commonly includes a description of the subject of the art, the desired medium (such as digital painting or photography), style (such as hyperrealistic or pop-art), lighting (such as rim lighting or crepuscular rays), color, and texture.^[51] Word order also affects the output of a text-to-image prompt. Words closer to the start of a prompt may be emphasized more heavily.^[6]

The Midjourney documentation encourages short, descriptive prompts: instead of "Show me a picture of lots of blooming California poppies, make them bright, vibrant orange, and draw them in an illustrated style with colored pencils", an effective prompt might be "Bright orange California poppies drawn with colored pencils".^[48]

Artist styles

Some text-to-image models are capable of imitating the style of particular artists by name. For example, the phrase in the style of Greg Rutkowski has been used in Stable Diffusion and Midjourney prompts to generate images in the distinctive style of Polish digital artist Greg Rutkowski.^[52] Famous artists such as Vincent van Gogh and Salvador Dalí have also been used for styling and testing.^[53]

Non-text prompts

Some approaches augment or replace natural language text prompts with non-text input.

Textual inversion and embeddings

For text-to-image models, textual inversion^[54] performs an optimization process to create a new word embedding based on a set of example images. This embedding vector acts as a "pseudo-word" which can be included in a prompt to express the content or style of the examples.

Image prompting

In 2023, Meta's AI research released Segment Anything, a computer vision model that can perform image segmentation by prompting. As an alternative to text prompts, Segment Anything can accept bounding boxes, segmentation masks, and foreground/background points.^[55]

Using gradient descent to search for prompts

In "prefix-tuning",^[56] "prompt tuning", or "soft prompting",^[57] floating-point-valued vectors are searched directly by gradient descent to maximize the log-likelihood on outputs.

Formally, let ${\displaystyle \mathbf {E} =\{\mathbf {e_{1}} ,\dots ,\mathbf {e_{k}} \}}$ be a set of soft prompt tokens (tunable embeddings), while ${\displaystyle \mathbf {X} =\{\mathbf {x_{1}} ,\dots ,\mathbf {x_{m}} \}}$ and ${\displaystyle \mathbf {Y} =\{\mathbf {y_{1}} ,\dots ,\mathbf {y_{n}} \}}$ be the token embeddings of the input and output respectively. During training, the tunable embeddings, input, and output tokens are concatenated into a single sequence ${\displaystyle {\text{concat}}(\mathbf {E}$ ;\mathbf {X} ;\mathbf {Y} )} , and fed to the LLMs. The losses are computed over the ${\displaystyle \mathbf {Y} }$ tokens; the gradients are backpropagated to prompt-specific parameters: in prefix-tuning, they are parameters associated with the prompt tokens at each layer; in prompt tuning, they are merely the soft tokens added to the vocabulary.^[58]

More formally, this is prompt tuning. Let an LLM be written as ${\displaystyle LLM(X)=F(E(X))}$, where ${\displaystyle X}$ is a sequence of linguistic tokens, ${\displaystyle E}$ is the token-to-vector function, and ${\displaystyle F}$ is the rest of the model. In prefix-tuning, one provides a set of input-output pairs ${\displaystyle \{(X^{i},Y^{i})\}_{i}}$, and then use gradient descent to search for ${\displaystyle \arg \max _{\tilde {Z}}\sum _{i}\log Pr[Y^{i}|{\tilde {Z}}\ast E(X^{i})]}$. In words, ${\displaystyle \log Pr[Y^{i}|{\tilde {Z}}\ast E(X^{i})]}$ is the log-likelihood of outputting ${\displaystyle Y^{i}}$, if the model first encodes the input ${\displaystyle X^{i}}$ into the vector ${\displaystyle E(X^{i})}$, then prepend the vector with the "prefix vector" ${\displaystyle {\tilde {Z}}}$, then apply ${\displaystyle F}$.

For prefix tuning, it is similar, but the "prefix vector" ${\displaystyle {\tilde {Z}}}$ is pre-appended to the hidden states in every layer of the model.

An earlier result^[59] uses the same idea of gradient descent search, but is designed for masked language models like BERT, and searches only over token sequences, rather than numerical vectors. Formally, it searches for ${\displaystyle \arg \max _{\tilde {X}}\sum _{i}\log Pr[Y^{i}|{\tilde {X}}\ast X^{i}]}$ where ${\displaystyle {\tilde {X}}}$ is ranges over token sequences of a specified length.

Prompt injection

Main article: Prompt injection

See also: SQL injection and Cross-site scripting

Prompt injection is a family of related computer security exploits carried out by getting a machine learning model (such as an LLM) which was trained to follow human-given instructions to follow instructions provided by a malicious user. This stands in contrast to the intended operation of instruction-following systems, wherein the ML model is intended only to follow trusted instructions (prompts) provided by the ML model's operator.^[60][61][62]

See also

• Social engineering (security)

References

1. Genkina, Dina (March 6, 2024). "AI Prompt Engineering is Dead: Long live AI prompt engineering". IEEE Spectrum. Retrieved January 18, 2025.
2. Radford, Alec; Wu, Jeffrey; Child, Rewon; Luan, David; Amodei, Dario; Sutskever, Ilya (2019). "Language Models are Unsupervised Multitask Learners" (PDF). OpenAI. We demonstrate language models can perform down-stream tasks in a zero-shot setting – without any parameter or architecture modification
3. Wahle, Jan Philip; Ruas, Terry; Xu, Yang; Gipp, Bela (2024). "Paraphrase Types Elicit Prompt Engineering Capabilities". In Al-Onaizan, Yaser; Bansal, Mohit; Chen, Yun-Nung (eds.). Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. Miami, Florida, USA: Association for Computational Linguistics. pp. 11004–11033. arXiv:2406.19898. doi:10.18653/v1/2024.emnlp-main.617.
4. Heaven, Will Douglas (April 6, 2022). "This horse-riding astronaut is a milestone on AI's long road towards understanding". MIT Technology Review. Retrieved August 14, 2023.
5. Wiggers, Kyle (June 12, 2023). "Meta open sources an AI-powered music generator". TechCrunch. Retrieved August 15, 2023. Next, I gave a more complicated prompt to attempt to throw MusicGen for a loop: "Lo-fi slow BPM electro chill with organic samples."
6. Diab, Mohamad; Herrera, Julian; Chernow, Bob (October 28, 2022). "Stable Diffusion Prompt Book" (PDF). Retrieved August 7, 2023. Prompt engineering is the process of structuring words that can be interpreted and understood by a text-to-image model. Think of it as the language you need to speak in order to tell an AI model what to draw.
7. "How to Write AI Photoshoot Prompts: A Guide for Better Product Photos". claid.ai. June 12, 2023. Retrieved June 12, 2023.
8. McCann, Bryan; Shirish, Nitish; Xiong, Caiming; Socher, Richard (2018). "The Natural Language Decathlon: Multitask Learning as Question Answering". arXiv:1806.08730 [cs.CL].
9. Knoth, Nils; Tolzin, Antonia; Janson, Andreas; Leimeister, Jan Marco (June 1, 2024). "AI literacy and its implications for prompt engineering strategies". Computers and Education: Artificial Intelligence. 6: 100225. doi:10.1016/j.caeai.2024.100225. ISSN 2666-920X.
10. Bach, Stephen H.; Sanh, Victor; Yong, Zheng-Xin; Webson, Albert; Raffel, Colin; Nayak, Nihal V.; Sharma, Abheesht; Kim, Taewoon; M Saiful Bari; Fevry, Thibault; Alyafeai, Zaid; Dey, Manan; Santilli, Andrea; Sun, Zhiqing; Ben-David, Srulik; Xu, Canwen; Chhablani, Gunjan; Wang, Han; Jason Alan Fries; Al-shaibani, Maged S.; Sharma, Shanya; Thakker, Urmish; Almubarak, Khalid; Tang, Xiangru; Radev, Dragomir; Mike Tian-Jian Jiang; Rush, Alexander M. (2022). "PromptSource: An Integrated Development Environment and Repository for Natural Language Prompts". arXiv:2202.01279 [cs.LG].
11. Wei, Jason; Wang, Xuezhi; Schuurmans, Dale; Bosma, Maarten; Ichter, Brian; Xia, Fei; Chi, Ed H.; Le, Quoc V.; Zhou, Denny (October 31, 2022). Chain-of-Thought Prompting Elicits Reasoning in Large Language Models. Advances in Neural Information Processing Systems (NeurIPS 2022). Vol. 35. arXiv:2201.11903.
12. Wei, Jason; Zhou (May 11, 2022). "Language Models Perform Reasoning via Chain of Thought". ai.googleblog.com. Retrieved March 10, 2023.
13. Chen, Brian X. (June 23, 2023). "How to Turn Your Chatbot Into a Life Coach". The New York Times.
14. Chen, Brian X. (May 25, 2023). "Get the Best From ChatGPT With These Golden Prompts". The New York Times. ISSN 0362-4331. Retrieved August 16, 2023.
15. Chen, Zijie; Zhang, Lichao; Weng, Fangsheng; Pan, Lili; Lan, Zhenzhong (June 16, 2024). "Tailored Visions: Enhancing Text-to-Image Generation with Personalized Prompt Rewriting". 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE. pp. 7727–7736. arXiv:2310.08129. doi:10.1109/cvpr52733.2024.00738. ISBN 979-8-3503-5300-6.
16. Sharan Narang and Aakanksha Chowdhery (April 4, 2022). "Pathways Language Model (PaLM): Scaling to 540 Billion Parameters for Breakthrough Performance".
17. Dang, Ekta (February 8, 2023). "Harnessing the power of GPT-3 in scientific research". VentureBeat. Retrieved March 10, 2023.
18. Montti, Roger (May 13, 2022). "Google's Chain of Thought Prompting Can Boost Today's Best Algorithms". Search Engine Journal. Retrieved March 10, 2023.
19. Chung, Hyung Won; Hou, Le; Longpre, Shayne; Zoph, Barret; Tay, Yi; Fedus, William; Li, Yunxuan; Wang, Xuezhi; Dehghani, Mostafa; Brahma, Siddhartha; Webson, Albert; Gu, Shixiang Shane; Dai, Zhuyun; Suzgun, Mirac; Chen, Xinyun; Chowdhery, Aakanksha; Castro-Ros, Alex; Pellat, Marie; Robinson, Kevin; Valter, Dasha; Narang, Sharan; Mishra, Gaurav; Yu, Adams; Zhao, Vincent; Huang, Yanping; Dai, Andrew; Yu, Hongkun; Petrov, Slav; Chi, Ed H.; Dean, Jeff; Devlin, Jacob; Roberts, Adam; Zhou, Denny; Le, Quoc V.; Wei, Jason (2022). "Scaling Instruction-Finetuned Language Models". arXiv:2210.11416 [cs.LG].
20. Wei, Jason; Tay, Yi (November 29, 2022). "Better Language Models Without Massive Compute". ai.googleblog.com. Retrieved March 10, 2023.
21. Kojima, Takeshi; Shixiang Shane Gu; Reid, Machel; Matsuo, Yutaka; Iwasawa, Yusuke (2022). "Large Language Models are Zero-Shot Reasoners". arXiv:2205.11916 [cs.CL].
22. Dickson, Ben (August 30, 2022). "LLMs have not learned our language — we're trying to learn theirs". VentureBeat. Retrieved March 10, 2023.
23. Garg, Shivam; Tsipras, Dimitris; Liang, Percy; Valiant, Gregory (2022). "What Can Transformers Learn In-Context? A Case Study of Simple Function Classes". arXiv:2208.01066 [cs.CL].
24. Brown, Tom; Mann, Benjamin; Ryder, Nick; Subbiah, Melanie; Kaplan, Jared D.; Dhariwal, Prafulla; Neelakantan, Arvind (2020). "Language models are few-shot learners". Advances in Neural Information Processing Systems. 33: 1877–1901. arXiv:2005.14165.
25. Wei, Jason; Tay, Yi; Bommasani, Rishi; Raffel, Colin; Zoph, Barret; Borgeaud, Sebastian; Yogatama, Dani; Bosma, Maarten; Zhou, Denny; Metzler, Donald; Chi, Ed H.; Hashimoto, Tatsunori; Vinyals, Oriol; Liang, Percy; Dean, Jeff; Fedus, William (August 31, 2022). "Emergent Abilities of Large Language Models". arXiv:2206.07682 [cs.CL]. In prompting, a pre-trained language model is given a prompt (e.g. a natural language instruction) of a task and completes the response without any further training or gradient updates to its parameters... The ability to perform a task via few-shot prompting is emergent when a model has random performance until a certain scale, after which performance increases to well-above random
26. Caballero, Ethan; Gupta, Kshitij; Rish, Irina; Krueger, David (July 24, 2023), Broken Neural Scaling Laws, arXiv, doi:10.48550/arXiv.2210.14891, arXiv:2210.14891, retrieved January 27, 2025
27. Wei, Jason; Tay, Yi; Bommasani, Rishi; Raffel, Colin; Zoph, Barret; Borgeaud, Sebastian; Yogatama, Dani; Bosma, Maarten; Zhou, Denny; Metzler, Donald; Chi, Ed H.; Hashimoto, Tatsunori; Vinyals, Oriol; Liang, Percy; Dean, Jeff; Fedus, William (August 31, 2022). "Emergent Abilities of Large Language Models". arXiv:2206.07682 [cs.CL].
28. Musser, George. "How AI Knows Things No One Told It". Scientific American. Retrieved May 17, 2023. By the time you type a query into ChatGPT, the network should be fixed; unlike humans, it should not continue to learn. So it came as a surprise that LLMs do, in fact, learn from their users' prompts—an ability known as in-context learning.
29. "What Can Transformers Learn In-Context? A Case Study of Simple Function Classes". NeurIPS. 2022. arXiv:2208.01066. Training a model to perform in-context learning can be viewed as an instance of the more general learning-to-learn or meta-learning paradigm
30. Wang, Xuezhi; Wei, Jason; Schuurmans, Dale; Le, Quoc; Chi, Ed; Narang, Sharan; Chowdhery, Aakanksha; Zhou, Denny (March 1, 2022). "Self-Consistency Improves Chain of Thought Reasoning in Language Models". arXiv:2203.11171 [cs.CL].
31. Diao, Shizhe; Wang, Pengcheng; Lin, Yong; Zhang, Tong (February 1, 2023). "Active Prompting with Chain-of-Thought for Large Language Models". arXiv:2302.12246 [cs.CL].
32. Yao, Shunyu; Yu, Dian; Zhao, Jeffrey; Shafran, Izhak; Griffiths, Thomas L.; Cao, Yuan; Narasimhan, Karthik (May 17, 2023). "Tree of Thoughts: Deliberate Problem Solving with Large Language Models". arXiv:2305.10601 [cs.CL].
33. OpenAI and over 200 people (March 27, 2023). "GPT-4 Technical Report". arXiv:2303.08774 [cs.CL]. [See Figure 8.]
34. Sclar, Melanie; Choi, Yejin; Tsvetkov, Yulia; Suhr, Alane (July 1, 2024). "Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting". arXiv:2310.11324 [cs.CL].
35. Leidinger, Alina; van Rooij, Robert; Shutova, Ekaterina (2023). Bouamor, Houda; Pino, Juan; Bali, Kalika (eds.). "The language of prompting: What linguistic properties make a prompt successful?". Findings of the Association for Computational Linguistics: EMNLP 2023. Singapore: Association for Computational Linguistics: 9210–9232. arXiv:2311.01967. doi:10.18653/v1/2023.findings-emnlp.618.
36. Linzbach, Stephan; Dimitrov, Dimitar; Kallmeyer, Laura; Evang, Kilian; Jabeen, Hajira; Dietze, Stefan (June 2024). "Dissecting Paraphrases: The Impact of Prompt Syntax and supplementary Information on Knowledge Retrieval from Pretrained Language Models". In Duh, Kevin; Gomez, Helena; Bethard, Steven (eds.). Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). Mexico City, Mexico: Association for Computational Linguistics. pp. 3645–3655. arXiv:2404.01992. doi:10.18653/v1/2024.naacl-long.201.
37. Polo, Felipe Maia; Xu, Ronald; Weber, Lucas; Silva, Mírian; Bhardwaj, Onkar; Choshen, Leshem; de Oliveira, Allysson Flavio Melo; Sun, Yuekai; Yurochkin, Mikhail (October 30, 2024). "Efficient multi-prompt evaluation of LLMs". arXiv:2405.17202 [cs.CL].
38. "How Each Index Works - LlamaIndex 🦙 v0.10.17". docs.llamaindex.ai. Retrieved April 8, 2024.
39. Lewis, Patrick; Perez, Ethan; Piktus, Aleksandra; Petroni, Fabio; Karpukhin, Vladimir; Goyal, Naman; Küttler, Heinrich; Lewis, Mike; Yih, Wen-tau; Rocktäschel, Tim; Riedel, Sebastian; Kiela, Douwe (2020). "Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks". Advances in Neural Information Processing Systems. 33. Curran Associates, Inc.: 9459–9474. arXiv:2005.11401.
40. Larson, Jonathan; Truitt, Steven (February 13, 2024), GraphRAG: Unlocking LLM discovery on narrative private data, Microsoft
41. Edge, Darren; Trinh, Ha; Cheng, Newman; Bradley, Joshua; Chao, Alex; Mody, Apurva; Truitt, Steven; Larson, Jonathan (2024). "From Local to Global: A Graph RAG Approach to Query-Focused Summarization". arXiv:2404.16130 [cs.CL].
42. Sequeda, Juan; Allemang, Dean; Jacob, Bryon (2023). "A Benchmark to Understand the Role of Knowledge Graphs on Large Language Model's Accuracy for Question Answering on Enterprise SQL Databases". arXiv:2311.07509 [cs.AI].
43. Singh, Chandan; Morris, John; Aneja, Jyoti; Rush, Alexander; Gao, Jianfeng (October 4, 2022). "Explaining Patterns in Data with Language Models via Interpretable Autoprompting". arXiv:2210.01848 [cs.LG].
44. Zhou, Yongchao; Ioan Muresanu, Andrei; Han, Ziwen; Paster, Keiran; Pitis, Silviu; Chan, Harris; Ba, Jimmy (November 1, 2022). "Large Language Models Are Human-Level Prompt Engineers". arXiv:2211.01910 [cs.LG].
45. Pryzant, Reid; Iter, Dan; Li, Jerry; Lee, Yin Tat; Zhu, Chenguang; Zeng, Michael (2023). "Automatic Prompt Optimization with "Gradient Descent" and Beam Search". Microsoft Azure AI. arXiv:2305.03495.
46. Zhang, Zhuosheng; Zhang, Aston; Li, Mu; Smola, Alex (October 1, 2022). "Automatic Chain of Thought Prompting in Large Language Models". arXiv:2210.03493 [cs.CL].
47. Monge, Jim Clyde (August 25, 2022). "Dall-E2 VS Stable Diffusion: Same Prompt, Different Results". MLearning.ai. Retrieved August 31, 2022.
48. "Prompts". docs.midjourney.com. Retrieved August 14, 2023.
49. Max Woolf (November 28, 2022). "Stable Diffusion 2.0 and the Importance of Negative Prompts for Good Results". Retrieved August 14, 2023.
50. Goldblum, R.; Pillarisetty, R.; Dauphinee, M. J.; Talal, N. (1975). "Acceleration of autoimmunity in NZB/NZW F1 mice by graft-versus-host disease". Clinical and Experimental Immunology. 19 (2): 377–385. ISSN 0009-9104. PMC 1538084. PMID 2403.
51. "Stable Diffusion prompt: a definitive guide". May 14, 2023. Retrieved August 14, 2023.
52. Heikkilä, Melissa (September 16, 2022). "This Artist Is Dominating AI-Generated Art and He's Not Happy About It". MIT Technology Review. Retrieved August 14, 2023.
53. Solomon, Tessa (August 28, 2024). "The AI-Powered Ask Dalí and Hello Vincent Installations Raise Uncomfortable Questions about Ventriloquizing the Dead". ARTnews.com. Retrieved January 10, 2025.
54. Gal, Rinon; Alaluf, Yuval; Atzmon, Yuval; Patashnik, Or; Bermano, Amit H.; Chechik, Gal; Cohen-Or, Daniel (2022). "An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion". arXiv:2208.01618 [cs.CV]. Using only 3-5 images of a user-provided concept, like an object or a style, we learn to represent it through new "words" in the embedding space of a frozen text-to-image model.
55. Kirillov, Alexander; Mintun, Eric; Ravi, Nikhila; Mao, Hanzi; Rolland, Chloe; Gustafson, Laura; Xiao, Tete; Whitehead, Spencer; Berg, Alexander C.; Lo, Wan-Yen; Dollár, Piotr; Girshick, Ross (April 1, 2023). "Segment Anything". arXiv:2304.02643 [cs.CV].
56. Li, Xiang Lisa; Liang, Percy (2021). "Prefix-Tuning: Optimizing Continuous Prompts for Generation". Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). pp. 4582–4597. doi:10.18653/V1/2021.ACL-LONG.353. S2CID 230433941. In this paper, we propose prefix-tuning, a lightweight alternative to fine-tuning... Prefix-tuning draws inspiration from prompting
57. Lester, Brian; Al-Rfou, Rami; Constant, Noah (2021). "The Power of Scale for Parameter-Efficient Prompt Tuning". Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. pp. 3045–3059. arXiv:2104.08691. doi:10.18653/V1/2021.EMNLP-MAIN.243. S2CID 233296808. In this work, we explore "prompt tuning," a simple yet effective mechanism for learning "soft prompts"...Unlike the discrete text prompts used by GPT-3, soft prompts are learned through back-propagation
58. Sun, Simeng; Liu, Yang; Iter, Dan; Zhu, Chenguang; Iyyer, Mohit (2023). "How Does In-Context Learning Help Prompt Tuning?". arXiv:2302.11521 [cs.CL].
59. Shin, Taylor; Razeghi, Yasaman; Logan IV, Robert L.; Wallace, Eric; Singh, Sameer (November 2020). "AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Prompts". Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). Online: Association for Computational Linguistics. pp. 4222–4235. doi:10.18653/v1/2020.emnlp-main.346. S2CID 226222232.
60. Willison, Simon (September 12, 2022). "Prompt injection attacks against GPT-3". simonwillison.net. Retrieved February 9, 2023.
61. Papp, Donald (September 17, 2022). "What's Old Is New Again: GPT-3 Prompt Injection Attack Affects AI". Hackaday. Retrieved February 9, 2023.
62. Vigliarolo, Brandon (September 19, 2022). "GPT-3 'prompt injection' attack causes bot bad manners". The Register. Retrieved February 9, 2023.