Abstract
Some retrieval-augmented generation systems can be limited by a reliance on static, predefined resource lists, which may become outdated, or by computationally expensive and potentially non-deterministic search queries. The disclosed technology describes a logic-based method for the dynamic determination and retrieval of supplemental information sources. A system, such as a client computing device or server, can parse an input to extract contextual identifiers and apply a configurable set of rules, patterns, and heuristics to infer the location of relevant resources. These resources may then be fetched to augment a large language model or other generative model. This approach can provide access to relevant, up-to-date information while potentially reducing the need for frequent manual updates or broad searches, offering a predictable and adaptable process relative to some conventional retrieval methods.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Foulds, Iain, "Logic-Based Dynamic Resource Determination for Retrieval-Augmented Generation", Technical Disclosure Commons, (December 11, 2025)
https://www.tdcommons.org/dpubs_series/9017