Follow the incentive: Who has an interest in presenting a gating network as fundamentally opposed to basic venture capital prediction methods?
This simplification masks the potential for integration, where such a network could refine existing predictions rather than contradict them.
A Swiss foundation, for example, could use a large language model (LLM) to extract textual information, then a gating network to weight these results with structured financial data, enhancing the robustness of decisions without rejecting the baseline method.
The motivation behind this dichotomy could be to position a technology as a replacement, rather than an improvement.