Why answer exposure, source influence, attribution accuracy, and claim support require separate metrics even when they involve the same citation.
Prominence measures observable exposure
Prominence includes position, word share, repetition, attributed answer share, and position-adjusted exposure. Aggarwal and colleagues use answer-level word-count measures to capture how visibly an already-retrieved source appears.
Absorption asks how deeply a source shapes the answer
A citation may be selected but contribute little to the generated text. Absorption estimates the degree to which a source supplies facts, phrasing, evidence, reasoning, or answer structure. It is often a proxy because model internals are not directly observable.
Selection Rate and Absorption Score belong side by side
Selection Rate describes how often a source is chosen. Absorption Score describes influence conditional on access or selection. A frequently cited but shallowly used source and a rarely cited but deeply used source create different strategic problems.
Fidelity and groundedness evaluate correctness
Fidelity asks whether the answer accurately attributes information and represents the source or brand. Groundedness asks whether generated claims are actually supported by cited evidence. Neither can be inferred from citation count, position, or sentiment.
Visibility without fidelity can amplify the wrong message.
Questions about this topic
What is prominence in a generated answer?+
The observable amount, position, and repetition of exposure received by a source or brand.
How is absorption different from citation?+
Citation records explicit selection; absorption estimates how much the source actually contributed to the answer.
Is absorption directly observable?+
Often not. It should be presented as a clearly defined proxy unless internal attribution evidence is available.
What does groundedness measure?+
Whether answer claims are supported by the evidence in the cited sources.
Can a prominent answer be low fidelity?+
Yes. A brand can receive extensive exposure while being inaccurately described or supported by mismatched citations.
References
Sources are listed in APA 7 style. Preprints are identified as such and should not be treated as peer-reviewed findings unless separately published.
- Aggarwal, P., Murahari, V., Rajpurohit, T., Kalyan, A., & Narasimhan, K. (2024). GEO: Generative engine optimization. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (pp. 5–16). Association for Computing Machinery. https://doi.org/10.1145/3637528.3671900
- Zhang, K., He, X., & Yao, J. (2026). From citation selection to citation absorption: A measurement framework for generative engine optimization across AI search platforms [Preprint]. arXiv. https://doi.org/10.48550/arXiv.2604.25707
- Martinez, O. (2026). Optimizing visibility in generative engines: A critical survey of generative engine optimization (2023–2026) [Preprint]. arXiv. https://doi.org/10.48550/arXiv.2607.14035