Liu, Jiacheng; Liu, Alisa; Lu, Ximing; Welleck, Sean; West, Peter; Le Bras, Ronan; Choi, Yejin; Hajishirzi, Hannaneh (May 2022). «Generated Knowledge Prompting for Commonsense Reasoning». Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (Dublin, Ireland: Association for Computational Linguistics): 3154-3169. arXiv:2110.08387. doi:10.18653/v1/2022.acl-long.225.
«Mesa-Optimization». Consultado el 17 de mayo de 2023. «Mesa-Optimization is the situation that occurs when a learned model (such as a neural network) is itself an optimizer.»
Wei, Jason; Tay, Yi; Bommasani, Rishi; Raffel, Colin; Zoph, Barret; Borgeaud, Sebastian; Yogatama, Dani; Bosma, Maarten et ál. (2022-08-31). «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"».
Caballero, Ethan; Gupta, Kshitij; Rish, Irina; Krueger, David (2022). "Broken Neural Scaling Laws". International Conference on Learning Representations (ICLR), 2023.
Wei, Jason; Tay, Yi; Bommasani, Rishi; Raffel, Colin; Zoph, Barret; Borgeaud, Sebastian; Yogatama, Dani; Bosma, Maarten et ál. (2022-08-31). «Emergent Abilities of Large Language Models». arXiv:2206.07682 [cs.CL].
Wei, Jason; Wang, Xuezhi; Schuurmans, Dale; Bosma, Maarten; Ichter, Brian; Xia, Fei; Chi, Ed H.; Le, Quoc V. et ál. (2022-10-31). «Chain-of-Thought Prompting Elicits Reasoning in Large Language Models». arXiv:2201.11903 [cs.CL].
«Transformers learn in-context by gradient descent». arXiv:2212.07677. «"Thus we show how trained Transformers become mesa-optimizers i.e. learn models by gradient descent in their forward pass"».
Garg, Shivam; Tsipras, Dimitris; Liang, Percy. «What Can Transformers Learn In-Context? A Case Study of Simple Function Classes». 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"».
Kojima, Takeshi; Shixiang Shane Gu; Reid, Machel; Matsuo, Yutaka; Iwasawa, Yusuke (2022). «Large Language Models are Zero-Shot Reasoners». arXiv:2205.11916 [cs.CL].
Liu, Jiacheng; Liu, Alisa; Lu, Ximing; Welleck, Sean; West, Peter; Le Bras, Ronan; Choi, Yejin; Hajishirzi, Hannaneh (May 2022). «Generated Knowledge Prompting for Commonsense Reasoning». Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (Dublin, Ireland: Association for Computational Linguistics): 3154-3169. arXiv:2110.08387. doi:10.18653/v1/2022.acl-long.225.
Zhou, Denny; Schärli (2022-05-01). «Least-to-Most Prompting Enables Complex Reasoning in Large Language Models». arXiv:2205.10625.
Lewis, Patrick; Perez, Ethan; Piktus, Aleksandra; Petroni, Fabio; Karpukhin, Vladimir; Goyal, Naman; Küttler, Heinrich; Lewis, Mike et al. (2020). «Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks». Advances in Neural Information Processing Systems (Curran Associates, Inc.) 33: 9459-9474. arXiv:2005.11401.Se sugiere usar |número-autores= (ayuda)
Fernando, Chrisantha; Banarse, Dylan; Michalewski, Henryk; Osindero, Simon; Rocktäschel, Tim (2023). Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution. arXiv:2309.16797.
Pryzant, Reid; Iter, Dan; Li, Jerry; Lee, Yin Tat; Zhu, Chenguang; Zeng, Michael (2023). Automatic Prompt Optimization with "Gradient Descent" and Beam Search. arXiv:2305.03495.
Guo, Qingyan; Wang, Rui; Guo, Junliang; Li, Bei; Song, Kaitao; Tan, Xu; Liu, Guoqing; Bian, Jiang et al. (2023). Connecting Large Language Models with Evolutionary Algorithms Yields Powerful Prompt Optimizers. arXiv:2309.08532.Se sugiere usar |número-autores= (ayuda)
Zhou, Yongchao; Ioan Muresanu, Andrei; Han, Ziwen; Paster, Keiran; Pitis, Silviu; Chan, Harris; Ba, Jimmy (2022-11-01). «Large Language Models Are Human-Level Prompt Engineers». arXiv:2211.01910 [cs.LG].
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."».
Sun, Simeng; Liu, Yang; Iter, Dan. «How Does In-Context Learning Help Prompt Tuning?». arXiv:2302.11521.
Greshake, Kai; Abdelnabi, Sahar; Mishra, Shailesh; Endres, Christoph; Holz, Thorsten; Fritz, Mario (2023-02-01). «Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection». arXiv:2302.12173 [cs.CR].
Perez, Fábio; Ribeiro, Ian (2022). «Ignore Previous Prompt: Attack Techniques For Language Models». arXiv:2211.09527 [cs.CL].
Branch, Hezekiah J.; Cefalu, Jonathan Rodriguez (2022). «Evaluating the Susceptibility of Pre-Trained Language Models via Handcrafted Adversarial Examples». arXiv:2209.02128 [cs.CL].
McAuliffe, Zachary. «Google's Latest AI Model Can Be Taught How to Solve Problems». CNET(en inglés). Consultado el 10 de marzo de 2023. «'Chain-of-thought prompting allows us to describe multistep problems as a series of intermediate steps,' Google CEO Sundar Pichai».
Liu, Jiacheng; Liu, Alisa; Lu, Ximing; Welleck, Sean; West, Peter; Le Bras, Ronan; Choi, Yejin; Hajishirzi, Hannaneh (May 2022). «Generated Knowledge Prompting for Commonsense Reasoning». Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (Dublin, Ireland: Association for Computational Linguistics): 3154-3169. arXiv:2110.08387. doi:10.18653/v1/2022.acl-long.225.
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. «In this paper, we propose prefix-tuning, a lightweight alternative to fine-tuning... Prefix-tuning draws inspiration from prompting».
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. doi:10.18653/V1/2021.EMNLP-MAIN.243. «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».
Selvi, Jose (5 de diciembre de 2022). «Exploring Prompt Injection Attacks». research.nccgroup.com. «Prompt Injection is a new vulnerability that is affecting some AI/ML models and, in particular, certain types of language models using prompt-based learning».
Selvi, Jose (5 de diciembre de 2022). «Exploring Prompt Injection Attacks». NCC Group Research Blog(en inglés estadounidense). Consultado el 9 de febrero de 2023.
neurips.cc
proceedings.neurips.cc
Lewis, Patrick; Perez, Ethan; Piktus, Aleksandra; Petroni, Fabio; Karpukhin, Vladimir; Goyal, Naman; Küttler, Heinrich; Lewis, Mike et al. (2020). «Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks». Advances in Neural Information Processing Systems (Curran Associates, Inc.) 33: 9459-9474. arXiv:2005.11401.Se sugiere usar |número-autores= (ayuda)
Radford, Alec (2019). «Language Models are Unsupervised Multitask Learners». OpenAI blog. «We demonstrate language models can perform down-stream tasks in a zero-shot setting – without any parameter or architecture modification».
openai.com
OpenAI (30 de noviembre de 2022). «Introducing ChatGPT». OpenAI Blog. Consultado el 16 de agosto de 2023. «what is the fermat's little theorem».
Diab, Mohamad (28 de octubre de 2022). «Stable Diffusion Prompt Book». Consultado el 7 de agosto de 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.»
Musser, George. «How AI Knows Things No One Told It». Scientific American. Consultado el 17 de mayo de 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.»
Willison, Simon (12 de septiembre de 2022). «Prompt injection attacks against GPT-3». simonwillison.net(en inglés británico). Consultado el 9 de febrero de 2023.
Wiggers, Kyle (12 de junio de 2023). «Meta open sources an AI-powered music generator». TechCrunch. Consultado el 15 de agosto de 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."».
Robinson, Reid (3 de agosto de 2023). «How to write an effective GPT-3 or GPT-4 prompt». Zapier. Consultado el 14 de agosto de 2023. «"Basic prompt: 'Write a poem about leaves falling.' Better prompt: 'Write a poem in the style of Edgar Allan Poe about leaves falling.'».
