
El problema de medir la memoria de los agentes de IA (y por qué lo estamos haciendo mal)
arXiv:2606.17328v1 Announce Type: new Abstract: LLM agents increasingly maintain long-term memory of user facts across sessions. Yet such memory is usually evaluated by aggregating accuracy over question rows or episodes. Because this approach scores question rows independently, even when several q
17/06/2026

















