Princeton University’s 2024 KDD Conference Paper GEO English Version

The paper “GEO: Generative Engine Optimization” at the 2024 KDD conference at Princeton University. Today, GEO Classroom (en.xxkt. org. cn) brings the English version of Princeton University’s 2024 GEO paper. I hope it is helpful to everyone.GEO: Generative Engine Optimization

ABSTRACT

The advent of large language models (LLMs) has ushered in a new paradigm of search engines that use generative models to gather and summarize information to answer user queries. This emerging technology, which we formalize under the unified framework of generative engines (GEs), can generate accurate and personalized responses, rapidly replacing traditional search engines like Google and Bing. Generative Engines typically satisfy queries by synthesizing information from multiple sources and summarizing them using LLMs.

CCS CONCEPTS

• Computing methodologies → Natural language processing; Machine learning; • Information systems → Web searching and information discovery.

KEYWORDS

generative models, search engines, datasets and benchmarks ACM Reference Format:

1 INTRODUCTION

The invention of traditional search engines three decades ago revolutionized information access and dissemination globally [4]. While they were powerful and ushered in a host of applications like academic research and e-commerce, they were limited to providing a list of relevant websites for user queries. However, the recent success of large language models [5, 21] has paved the way for better systems like BingChat, Google’s SGE, and perplexity.ai that combine conventional search engines with generative models. We dub these systems generative engines (GE) because they search for information and generate multi-modal responses by using multiple sources.Figure 2: Overview of Generative Engines. Generative Engines primrarily consists of a set of generative models and a search engine to retrieve relevant documents. Generative Engines take user query as input and through a series of steps generate a final response that is grounded in the retrieved sources with inline attributions.

2 FORMULATION & METHODOLOGY

2.1 Formulation of Generative EnginesDespite the deployment of numerous generative engines to millions of users, there is currently no standard framework. We provide a formulation that accommodates various modular components in their design. We describe a generative engine, which includes several backend generative models and a search engine for source retrieval.Generative Engines comprise two crucial components: a.) A set of generative models𝐺 = {𝐺1,𝐺2…𝐺𝑛}, each serving a specific purpose like query reformulation or summarization, and b.) A search engine 𝑆𝐸 that returns a set of sources 𝑆 = {𝑠1, 𝑠2…𝑠𝑚} given a query 𝑞.2.2 Generative Engine OptimizationThe advent of search engines led to search engine optimization(SEO), a process to help website creators optimize their content to improve search engine rankings. Higher rankings correlate with increased visibility and website traffic. However, traditional SEO methods are not directly applicable to Generative Engines. This is because, unlike traditional search engines, the generative model in generative engines is not limited to keyword matching, and the use of language models in ingesting source documents and response generation results in a more nuanced understanding of text documents and user query. With generative engines rapidly emerging as the primary information delivery paradigm and SEO is not directly applicable; new techniques are needed. To this end,we propose Generative Engine Optimization, a new paradigm where content creators aim to increase their visibility (or impression) in generative engine responses. We define the visibility of a website (also referred to as a citation)𝑐𝑖 in a cited response 𝑟 by the function 𝐼𝑚𝑝(𝑐𝑖, 𝑟), which the website creator wants to maximize.measures the relevance of citation 𝑐𝑖 to the query 𝑞 in the context of response 𝑟 and 𝑓 is determined by the exact algorithmic design of generative engine and is a black-box function to end-users. Further, both the functions 𝐼𝑚𝑝 and 𝑅𝑒𝑙 are subjective and not well-defined yet for generative engines, and we define them next.2.2.1 Impressions for Generative Engines.In SEO, a website’s impression (or visibility) is determined by its average ranking over a range of queries. However, generative engines’ output nature necessitates different impression metrics. Unlike search engines,Generative Engines combine information from multiple sources in a single response. Factors such as length, uniqueness, and presentation of the cited website determine the true visibility of a citation.Here 𝑆𝑐𝑖 is the set of sentences citing 𝑐𝑖, 𝑆𝑟 is the set of sentences in the response, and |𝑠| is the number of words in sentence 𝑠. In cases where a sentence is cited by multiple sources, we share the word count equally with all the citations. Intuitively, a higher word count correlates with the source playing a more important part in the answer, and thus, the user gets higher exposure to that source.However, since “Word Count” is not impacted by the ranking of the citations (whether it appears first, for example), we propose a position-adjusted count that reduces the weight by an exponentially decaying function of the citation position:Intuitively, sentences that appear first in the response are more3 EXPERIMENTAL SETUP3.3 GEO MethodsTable 1: Absolute impression metrics of GEO methods on GEO-bench. Performance Measured on Two metrics and theirThe modified response 𝑟

4 RESULTS

We evaluate various Generative Engine Optimization methodsTable 3: Top Performing categories for each of the GEO methods. Website-owners can choose relevant GEO strategy basedFurther, given generative models are often designed to follow instructions, one would expect a more persuasive and authoritative tone in website content to boost visibility. However, we find no significant improvement, demonstrating that Generative Engines are already somewhat robust to such changes. This highlights the need for website owners to focus on improving content presentation and credibility.

5 ANALYSIS

5.1 Domain-Specific Generative Engine OptimizationsIn Section 4, we presented the improvements achieved by GEO across the entirety of the GEO-bench benchmark. However, in

5.2 Optimization of Multiple WebsitesIn the evolving landscape of Generative Engines, GEO methods are expected to become widely adopted, leading to a scenario where all source contents are optimized using GEO. To understand the implications, we conducted an evaluation of GEO methods by optimizing all source contents simultaneously, with results presented in Table 2. A key observation is the differential impact of GEO on websites based on their Search Engine Results Pages (SERP) ranking.

5.3 Combination of GEO StrategiesWhile individual GEO strategies show significant improvements across various domains, in practice, website owners are expected to employ multiple strategies in conjunction. To study the performance improvements achieved by combining GEO strategies, we consider all pairs of combinations of the top 4 performing GEO methods, namely Cite Sources, Fluency Optimization, Statistics Addition, and Quotation Addition. Figure 4 displays the heatmap of relative improvement in the Position-Adjusted Word Count visibility metric achieved by combining different GEO strategies.

6 GEO IN THE WILD

GEO IN THE WILD : EXPERIMENTS WITH DEPLOYED GENERATIVE ENGINETo reinforce the efficacy of our proposed Generative Engine Optimization methods, we evaluate them on Perplexity.ai, a real deployed Generative Engine with a large user base. Results are in Table 5. Similar to our generative engine, Quotation Addition performs best in Position-Adjusted Word Count with a 22% improvement over the baseline.

7 RELATED WORK

Evidence-based Answer Generation: Previous works have used several techniques for answer generation backed by sources. Nakano et al. [19] trained GPT-3 to navigate web environments to generate source-backed answers.

8 CONCLUSION

In this work, we formulate search engines augmented with generative models that we dub generative engines. We propose Generative Engine Optimization (GEO) to empower content creators  to optimize their content under generative engines.

9 LIMITATIONS

While we rigorously test our proposed methods on two generative engines, including a publicly available one, methods may need to adapt over time as GEs evolve, mirroring the evolution of SEO. Additionally, despite our efforts to ensure the queries in our GEObench closely resemble real-world queries, the nature of queries can change over time, necessitating continuous updates.

10 ACKNOWLEDGEMENTS

This material is based upon work supported by the National Science Foundation under Grant No. 2107048. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.A CONVERSATIONAL GENERATIVE ENGINE In Section 2.1, we discussed a single-turn Generative Enginethat outputs a single response given the user query. However, one of the strengths of upcoming Generative Engines will be their ability to engage in an active back-and-forth conversation with the user.where 𝑡 is the turn number. Further, to engage the user in a conversation, a separate LLM, 𝐿𝑓 𝑜𝑙𝑙𝑜𝑤 or 𝐿𝑟𝑒𝑠𝑝 , may generate suggested follow-up queries based on 𝐻, 𝑃𝑈 , and 𝑟 𝑡+1 .• Difficulty Level: The complexity of the query, ranging from simple to complex.C.1 GEO in the Wild : Experiments with Deployed Generative Engine We also evaluate our proposed Generative Engine Optimization methods on real-world deployed Generative Engine: Perplexity.ai. Since perplexity.ai does not allow the user to specify source URLs, we instead provide source text as file uploads to perplexity.ai while ensuring all answers are generated only using the file sources provided. We evaluate all our methods on a subset of 200 samples of our test set. Results using Perplexity.ai are shown in Table 7.REFERENCES (References)[1] Daria Alexander, Wojciech Kusa, and Arjen P. de Vries. 2022. ORCAS-I: Queries Annotated with Intent using Weak Supervision. Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval (2022). https://api.semanticscholar.org/CorpusID:248495926 [2] Prashant Ankalkoti. 2017. Survey on Search Engine Optimization Tools & Techniques. Imperial journal of interdisciplinary research 3 (2017). https: //api.semanticscholar.org/CorpusID:116487363 [3] Akari Asai, Xinyan Velocity Yu, Jungo Kasai, and Hannaneh Hajishirzi. 2021. One Question Answering Model for Many Languages with Cross-lingual  Dense Passage Retrieval. In Neural Information Processing Systems. https: //api.semanticscholar.org/CorpusID:236428949 [4] Sergey Brin and Lawrence Page. 1998. The Anatomy of a Large-Scale Hypertextual Web Search Engine. Comput. Networks 30 (1998), 107–117. https: //api.semanticscholar.org/CorpusID:7587743 [5] Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel Ziegler, Jeffrey Wu, Clemens Winter, Chris Hesse, Mark Chen, Eric Sigler, Mateusz Litwin, Scott Gray, Benjamin Chess, Jack Clark, Christopher Berner, Sam McCandlish, Alec Radford, Ilya Sutskever, and Dario Amodei. 2020. Language Models are Few-Shot Learners. In Advances in Neural Information Processing Systems, H. Larochelle, M. Ranzato, R. Hadsell, M.F. Balcan, and H. Lin (Eds.), Vol. 33. Curran Associates, Inc., 1877–1901. https://proceedings.neurips.cc/paper_files/paper/2020/file/ 1457c0d6bfcb4967418bfb8ac142f64a-Paper.pdf [6] Nick Craswell, Bhaskar Mitra, Emine Yilmaz, Daniel Fernando Campos, and Jimmy J. Lin. 2021. MS MARCO: Benchmarking Ranking Models in the Large-Data Regime. Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval (2021). https://api.semanticscholar.org/ CorpusID:234336491 [7] Brian Dean. 2023. We Analyzed 4 Million Google Search Results. Here’s What We Learned About Organic Click Through Rate. https://backlinko.com/googlectr-stats Accessed: 2024-06-08. [8] Danny Goodwin. 2011. Top Google Result Gets 36.4% of Clicks [Study]. https://www.searchenginewatch.com/2011/04/21/top-google-resultgets-36-4-of-clicks-study/ [9] Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat, and Ming-Wei Chang. 2020. REALM: Retrieval-Augmented Language Model Pre-Training. ArXiv abs/2002.08909 (2020). https://api.semanticscholar.org/CorpusID:211204736 [10] Ziwei Ji, Nayeon Lee, Rita Frieske, Tiezheng Yu, Dan Su, Yan Xu, Etsuko Ishii, Ye Jin Bang, Andrea Madotto, and Pascale Fung. 2023. Survey of hallucination in natural language generation. Comput. Surveys 55, 12 (2023), 1–38. [11] Aounon Kumar and Himabindu Lakkaraju. 2024. Manipulating Large Language Models to Increase Product Visibility. arXiv:2404.07981 [cs.IR] [12] R.Anil Kumar, Zaiduddin Shaik, and Mohammed Furqan. 2019. A Survey on Search Engine Optimization Techniques. International Journal of P2P Network Trends and Technology (2019). https://doi.org/10.14445/22492615/IJPTT-V9I1P402 [13] Tom Kwiatkowski, Jennimaria Palomaki, Olivia Redfield, Michael Collins, Ankur P. Parikh, Chris Alberti, Danielle Epstein, Illia Polosukhin, Jacob Devlin, Kenton Lee, Kristina Toutanova, Llion Jones, Matthew Kelcey, Ming-Wei Chang, Andrew M. Dai, Jakob Uszkoreit, Quoc V. Le, and Slav Petrov. 2019. Natural Questions: A Benchmark for Question Answering Research. Transactions of the Association for Computational Linguistics 7 (2019), 453–466. https: //api.semanticscholar.org/CorpusID:86611921 [14] Nelson F. Liu, Tianyi Zhang, and Percy Liang. 2023. Evaluating Verifiability in Generative Search Engines. ArXiv abs/2304.09848 (2023). https://api. semanticscholar.org/CorpusID:258212854 [15] Yang Liu, Dan Iter, Yichong Xu, Shuo Wang, Ruochen Xu, and Chenguang Zhu. 2023. G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment. ArXiv abs/2303.16634 (2023). https://api.semanticscholar.org/CorpusID:257804696 [16] G. D. Maayan. 2023. How Google SGE will impact your traffic – and 3 SGE recovery case studies. Search Engine Land (5 Sep 2023). https://searchengineland.com/how-google-sge-will-impact-your-trafficand-3-sge-recovery-case-studies-431430 [17] Jacob Menick, Maja Trebacz, Vladimir Mikulik, John Aslanides, Francis Song, Martin Chadwick, Mia Glaese, Susannah Young, Lucy Campbell-Gillingham, Geoffrey Irving, and Nathan McAleese. 2022. Teaching language models to support answers with verified quotes. ArXiv abs/2203.11147 (2022). https: //api.semanticscholar.org/CorpusID:247594830 [18] Grégoire Mialon, Roberto Dessì, Maria Lomeli, Christoforos Nalmpantis, Ramakanth Pasunuru, Roberta Raileanu, Baptiste Rozi è re, Timo Schick, Jane Dwivedi-Yu, Asli Celikyilmaz, Edouard Grave, Yann LeCun, and Thomas Scialom. 2023. Augmented Language Models: a Survey. ArXiv abs/2302.07842 (2023). https://api.semanticscholar.org/CorpusID:256868474 [19] Reiichiro Nakano, Jacob Hilton, S. Arun Balaji, Jeff Wu, Ouyang Long, Christina Kim, Christopher Hesse, Shantanu Jain, Vineet Kosaraju, William Saunders, Xu Jiang, Karl Cobbe, Tyna Eloundou, Gretchen Krueger, Kevin Button, Matthew Knight, Benjamin Chess,and John Schulman. 2021. WebGPT: Browser-assisted question-answering with human feedback. ArXiv abs/2112.09332 (2021). https: //api.semanticscholar.org/CorpusID:245329531 [20] OpenAI. 2022. Introducing ChatGPT. https://openai.com/index/chatgpt/ [21] OpenAI, Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, Red Avila, Igor Babuschkin, Suchir Balaji, Valerie Balcom, Paul Baltescu, Haiming Bao, Mohammad Bavarian, Jeff Belgum, Irwan Bello, Jake Berdine, Gabriel Bernadett-Shapiro, Christopher Berner, Lenny Bogdonoff, Oleg Boiko, Madelaine Boyd, Anna-Luisa Brakman, Greg Brockman, Tim Brooks,Miles Brundage, Kevin Button, Trevor Cai, Rosie Campbell, Andrew Cann, Brittany Carey, Chelsea Carlson, Rory Carmichael, Brooke Chan, Che Chang, Fotis Chantzis, Derek Chen, Sully Chen, Ruby Chen, Jason Chen, Mark Chen, Ben Chess, Chester Cho, Casey Chu, Hyung Won Chung, Dave Cummings, Jeremiah Currier, Yunxing Dai, Cory Decareaux, Thomas Degry, Noah Deutsch, Damien Deville, Arka Dhar, David Dohan, Steve Dowling, Sheila Dunning, Adrien Ecoffet, Atty Eleti, Tyna Eloundou, David Farhi, Liam Fedus, Niko Felix, Sim ó n Posada Fishman, Juston Forte, Isabella Fulford, Leo Gao, Elie Georges, Christian Gibson, Vik Goel, Tarun Gogineni, Gabriel Goh, Rapha Gontijo-Lopes, Jonathan Gordon, Morgan Grafstein, Scott Gray, Ryan Greene, Joshua Gross, Shixiang Shane Gu, Yufei Guo, Chris Hallacy, Jesse Han, Jeff Harris, Yuchen He, Mike Heaton, Johannes Heidecke, Chris Hesse, Alan Hickey, Wade Hickey, Peter Hoeschele, Brandon Houghton, Kenny Hsu, Shengli Hu, Xin Hu, Joost Huizinga, Shantanu Jain, Shawn Jain, Joanne Jang, Angela Jiang, Roger Jiang, Haozhun Jin, Denny Jin, Shino Jomoto, Billie Jonn, Heewoo Jun, Tomer Kaftan, Łukasz Kaiser, Ali Kamali, Ingmar Kanitscheider, Nitish Shirish Keskar, Tabarak Khan, Logan Kilpatrick, Jong Wook Kim, Christina Kim, Yongjik Kim, Jan Hendrik Kirchner, Jamie Kiros, Matt Knight, Daniel Kokotajlo, Łukasz Kondraciuk, Andrew Kondrich, Aris Konstantinidis, Kyle Kosic, Gretchen Krueger, Vishal Kuo, Michael Lampe, Ikai Lan, Teddy Lee, Jan Leike, Jade Leung, Daniel Levy, Chak Ming Li, Rachel Lim, Molly Lin, Stephanie Lin, Mateusz Litwin, Theresa Lopez, Ryan Lowe, Patricia Lue, Anna Makanju, Kim Malfacini, Sam Manning, Todor Markov, Yaniv Markovski, Bianca Martin, Katie Mayer, Andrew Mayne, Bob McGrew, Scott Mayer McKinney, Christine McLeavey, Paul McMillan, Jake McNeil, David Medina, Aalok Mehta, Jacob Menick, Luke Metz, Andrey Mishchenko, Pamela Mishkin, Vinnie Monaco, Evan Morikawa, Daniel Mossing, Tong Mu, Mira Murati, Oleg Murk, David M é ly, Ashvin Nair, Reiichiro Nakano, Rajeev Nayak, Arvind Neelakantan, Richard Ngo, Hyeonwoo Noh, Long Ouyang, Cullen O’Keefe, Jakub Pachocki, Alex Paino, Joe Palermo, Ashley Pantuliano, Giambattista Parascandolo, Joel Parish, Emy Parparita, Alex Passos, Mikhail Pavlov, Andrew Peng, Adam Perelman, Filipe de Avila Belbute Peres, Michael Petrov, Henrique Ponde de Oliveira Pinto, Michael, Pokorny, Michelle Pokrass, Vitchyr H. Pong, Tolly Powell, Alethea Power, Boris Power, Elizabeth Proehl, Raul Puri, Alec Radford, Jack Rae, Aditya Ramesh, Cameron Raymond, Francis Real, Kendra Rimbach, Carl Ross, Bob Rotsted, Henri Roussez, Nick Ryder, Mario Saltarelli, Ted Sanders, Shibani Santurkar, Girish Sastry, Heather Schmidt, David Schnurr, John Schulman, Daniel Selsam, Kyla Sheppard, Toki Sherbakov, Jessica Shieh, Sarah Shoker, Pranav Shyam, Szymon Sidor, Eric Sigler, Maddie Simens, Jordan Sitkin, Katarina Slama, Ian Sohl, Benjamin Sokolowsky, Yang Song, Natalie Staudacher, Felipe Petroski Such, Natalie Summers, Ilya Sutskever, Jie Tang, Nikolas Tezak, Madeleine B. Thompson, Phil Tillet, Amin Tootoonchian, Elizabeth Tseng, Preston Tuggle, Nick Turley, Jerry Tworek, Juan Felipe Cer ó n Uribe, Andrea Vallone, Arun Vijayvergiya, Chelsea Voss, Carroll Wainwright, Justin Jay Wang, Alvin Wang, Ben Wang, Jonathan Ward, Jason Wei, CJ Weinmann, Akila Welihinda, Peter Welinder, Jiayi Weng, Lilian Weng, Matt Wiethoff, Dave Willner, Clemens Winter, Samuel Wolrich, Hannah Wong, Lauren Workman, Sherwin Wu, Jeff Wu, Michael Wu, Kai Xiao, Tao Xu, Sarah Yoo, Kevin Yu, Qiming Yuan, Wojciech Zaremba, Rowan Zellers, Chong Zhang, Marvin Zhang, Shengjia Zhao, Tianhao Zheng, Juntang Zhuang, William Zhuk, and Barret Zoph. 2024. GPT-4 Technical Report. arXiv:2303.08774 [cs.CL] [22] A. Shahzad, Deden Witarsyah Jacob, Nazri M. Nawi, Hairulnizam Bin Mahdin, and Marheni Eka Saputri. 2020. The new trend for search engine optimization, tools and techniques. Indonesian Journal of Electrical Engineering and Computer Science 18 (2020), 1568. https://api.semanticscholar.org/CorpusID:213123106 [23] Kurt Shuster, Jing Xu, Mojtaba Komeili, Da Ju, Eric Michael Smith, Stephen Roller, Megan Ung, Moya Chen, Kushal Arora, Joshua Lane, Morteza Behrooz, W.K.F. Ngan, Spencer Poff, Naman Goyal, Arthur Szlam, Y-Lan Boureau, Melanie Kambadur, and Jason Weston. 2022. BlenderBot 3: a deployed conversational agent that continually learns to responsibly engage. ArXiv abs/2208.03188 (2022). https://api.semanticscholar.org/CorpusID:251371589 [24] Romal Thoppilan, Daniel De Freitas, Jamie Hall, Noam Shazeer, Apoorv Kulshreshtha, Heng-Tze Cheng, Alicia Jin, Taylor Bos, Leslie Baker, Yu Du, YaGuang Li, Hongrae Lee, Huaixiu Steven Zheng, Amin Ghafouri, Marcelo Menegali, Yanping Huang, Maxim Krikun, Dmitry Lepikhin, James Qin, Dehao Chen,Yuanzhong Xu, Zhifeng Chen, Adam Roberts, Maarten Bosma, Vincent Zhao, Yanqi Zhou, Chung-Ching Chang, Igor Krivokon, Will Rusch, Marc Pickett, Pranesh Srinivasan, Laichee Man, Kathleen Meier-Hellstern, Meredith Ringel Morris, Tulsee Doshi, Renelito Delos Santos, Toju Duke, Johnny Soraker, Ben Zevenbergen, Vinodkumar Prabhakaran, Mark Diaz, Ben Hutchinson, Kristen Olson, Alejandra Molina, Erin Hoffman-John, Josh Lee, Lora Aroyo, Ravi Rajakumar, Alena Butryna, Matthew Lamm, Viktoriya Kuzmina, Joe Fenton, Aaron Cohen, Rachel Bernstein, Ray Kurzweil, Blaise Aguera-Arcas, Claire Cui, Marian Croak, Ed Chi, and Quoc Le. 2022. LaMDA: Language Models for Dialog Applications. arXiv:2201.08239 [cs.CL] [25] Chunting Zhou, Pengfei Liu, Puxin Xu, Srini Iyer, Jiao Sun, Yuning Mao, Xuezhe Ma, Avia Efrat, Ping Yu, L. Yu, Susan Zhang, Gargi Ghosh, Mike Lewis, Luke Zettlemoyer, and Omer Levy. 2023. LIMA: Less Is More for Alignment. ArXiv abs/2305.11206 (2023). https://api.semanticscholar.org/CorpusID:258822910The above is GEO Little Classroom Network（ https://en.xxkt.org.cn/ ）Here is the GEO English version of the paper presented at the 2024 KDD conference at Princeton University. Thank you for watching.

非特殊说明，本文为小小课堂SEO自学网原创，欢迎转载并保留版权 https://www.xxkt.org.cn/

本站提供SEO与GEO培训、咨询、诊断，微信（电话）：13722793092 微信公众号：xxktorg

标签：Generative Engine Optimization, GEO paper, Princeton University GEO Paper 文章最后更新时间：六月 18, 2026

一条回应：“Princeton University’s 2024 KDD Conference Paper GEO English Version”

1. 小小课堂SEO培训说道：

   2026年6月5日 上午2:05

   论文及源码地址

   https://generative-engines.com/GEO/

   回复