We cite these papers as the basis for our methods and evidence; each includes an arXiv link and BibTeX.
GEO: Generative Engine Optimization Aggarwal, P., Murahari, V., Rajpurohit, T., Kalyan, A., Narasimhan, K., Deshpande, A. (2023). arXiv preprint (KDD 2024). arXiv Defines GEO as optimizing content for generative engines and reports measurable visibility gains from citation-oriented content.
Evaluating Verifiability in Generative Search Engines Liu, N. F., Zhang, T., Liang, P. (2023). arXiv preprint. arXiv Shows that generative search answers often lack citations and can contradict their own sources — why evidence and citation checks matter.
Lost in the Middle: How Language Models Use Long Contexts Liu, N. F., Lin, K., Hewitt, J., Paranjape, A., Bevilacqua, M., Petroni, F., Liang, P. (2023). arXiv preprint (TACL 2024). arXiv Provides evidence that models use information in the middle of long contexts poorly — supports chunking content and putting key facts early.
Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., Küttler, H., Lewis, M., Yih, W., Rocktäschel, T., Riedel, S., Kiela, D. (2020). NeurIPS 2020. arXiv The foundational RAG paper: retrieval + generation is why machine-readable, retrievable content matters.
Retrieval-Augmented Generation for Large Language Models: A Survey Gao, Y., Xiong, Y., Gao, X., Jia, K., Pan, J., Bi, Y., Dai, Y., Sun, J., Wang, H. (2023). arXiv preprint. arXiv Survey of RAG architectures and chunking/retrieval trade-offs — useful for structuring content for retrieval.
WebGPT: Browser-assisted question-answering with human feedback Nakano, R., Hilton, J., Balaji, S., Wu, J., Ouyang, L., Kim, C., Hesse, C., Jain, S., Kosaraju, V., Saunders, W., Jiang, X., Cobbe, K., Ziegler, D., Myers, J., Plappert, T., Zaremba, W., Schulman, J., Hilton, J. (2021). arXiv preprint. arXiv Early browser-augmented answering work: models read and cite web pages, which is why page-level readability and citations matter.