LATIDIA · Investigación

¿Los agentes de IA entienden la arquitectura informática?

arXiv: 2609.19387v1Announce Type: new Abstract: A los agentes se les pide cada vez más que diseñen hardware y se informa cada vez más que tienen éxito. Dichos informes establecen que un diseño mejoró; no pueden establecer por qué. Un agente t

La noticia

arXiv:2609.19387v1 Announce Type: new Abstract: Agents are increasingly asked to design hardware, and increasingly reported to succeed. Such reports establish that a design improved; they cannot establish why. An agent that improves an accelerator may be reasoning about the machine, or may be searching competently over knobs whose meaning it never recovers -- and only the first transfers to the next architecture. Existing evaluations cannot tell the two apart, because they vary the agent while holding the framing of the problem fixed. We do the opposite. AutoTuring hands the same agent the same 15-dimensional accelerator space twice: once as named architectural knobs with simulator counters, once as anonymous variables on