LATIDIA · Ciberseguridad

VRMaaS: Modelo de recompensa verificable como servicio para la alineación fiable de los LLM

arXiv:2611.14984v1 Tipo de anuncio: nuevo Resumen: Los modelos de recompensa (RM) se implementan cada vez más como componentes de servicio en los procesos de alineación de los LLM, donde los flujos de trabajo de pago, auditoría, moderación y gobernanza actúan sobre las puntuaciones.

La noticia

arXiv:2611.14984v1 Announce Type: new Abstract: Reward models (RMs) are increasingly deployed as service components in LLM alignment pipelines, where payment, audit, moderation, and governance workflows act on the scores they return. An inference proof alone does not show whether a score belongs to the intended request, model checkpoint, numerical configuration, and verifier: in a controlled comparison that isolates the accepted statement from prover scale, a generic inference proof wrapped for on-chain checking remains valid after six classes of such service-field substitution. We formulate verifiable reward-model-as-a-service (VRMaaS) as a service-output integrity problem and present a framework in which each returned reward becomes a proof-carrying artifact. Its off-chain acceptance rule binds request