What Comes After Transformers?

The Data Exchange · Veröffentlicht am 24. Sept. 2026

Zuzanna Stamirowska calls RAG, context, and memory bottlenecks in 2023 and 2024 for what she describes as 'AI with no ceiling, if you wish.' She recalls realizing that reasoning should not just be linguistic, and says latent thinking can help solve constraint-satisfaction problems and math with 'tiny, tiny models.'. Lies 3 Standpunkte mit Belegen und Links zu den Originalquellen.

Ben LoricaZuzanna Stamirowska

Kernaussagen lesen Herausgeber öffnen

Auf einen Blick

RAG, context, and memory were bottlenecks in 2023–2024
Stamirowska says the team saw RAG, context, and memory as bottlenecks in 2023 and 2024. They were not efficient or sufficient for what she describes as 'AI with no ceiling, if you wish.'
Unterstützendes Moment lesen · Publisher transcript paragraph 7
Stamirowska says reasoning shouldn't just be linguistic
Stamirowska says, 'We realized that reasoning shouldn't just be linguistic.' She says verbalizing things in chain-of-thought is not optimal: it involves 'gluing it on top of a transformer,' and transformers are naturally ill-suited to dealing with time and learning from experience.
Unterstützendes Moment lesen · Publisher transcript paragraph 8
Latent thinking can help solve constraint-satisfaction and math problems
Stamirowska says latent thinking can help solve constraint-satisfaction problems and math with 'tiny, tiny models.'
Unterstützendes Moment lesen · Publisher transcript paragraph 37

Schlüsselmomente3

Kurze, zitierte Textstellen mit Kontext zur Überprüfung. Das vollständige Gespräch bleibt beim Herausgeber.

Publisher transcript paragraph 7

post-transformer architecture motivation

RAG, context, and memory were bottlenecks in 2023–2024

Originalauszug

But very concretely, what prompted us to start Pathway and to start working on post-transformer architectures was that we were looking forward to reasoning becoming a thing. We have a very strong background in dealing with the dimension of time in machine learning—any sort of online learning. And we saw RAG, context, and memory, and we’re talking about 2023 and 2024, as bottlenecks. They were not efficient or sufficient to unlock AI as we would like—AI with no ceiling, if you wish.

Zuzanna Stamirowska·24. Sept. 2026

Publisher transcript paragraph 8

reasoning modality

Stamirowska says reasoning shouldn't just be linguistic

Originalauszug

Specifically, we’re talking about continual learning, memory, and reasoning first. And as we moved through it, we realized that reasoning shouldn’t just be linguistic. Verbalizing things in chain of thought is not the optimal way to do reasoning. With chain of thought, we’re essentially gluing it on top of a transformer, and transformers are naturally ill-suited to dealing with time and learning from experience.

Zuzanna Stamirowska·24. Sept. 2026

Publisher transcript paragraph 37

latent reasoning benefits

Latent thinking can help solve constraint-satisfaction and math problems

Originalauszug

Latent thinking has huge benefits. It can help us solve constraint-satisfaction problems natively, like what we’re showing. We’re solving constraint-satisfaction problems and math with tiny, tiny models. And then there’s cost efficiency. There are just insane cost gains coming from latent reasoning.

Zuzanna Stamirowska·24. Sept. 2026

Quelle & Methodik

Diese Standpunkte sind mit ihren Originalquellen verknüpft. Paraphrasen sind gekennzeichnet und keine wörtlichen Zitate.

Transkript oder Quellenmaterial öffnen Ein Problem melden