5 papers
The Assistant as a Privileged Persona: A canonical reference in cross-persona self-recognition
Asvin G
Post-trained language models can recognize their own outputs from a sentence or two out of context. In a companion paper \citep{jack2026twomodes} we showed they can also recognize…
From Simulation to Enaction: Post-trained language models recognize and react to their own generations
Asvin G., Jack Lindsey
Language models are pretrained as passive predictors with no incentive to model the consequences of their own outputs. Post-training changes this: a model producing its own respons…
Matrix Points on Varieties
Asvin G, Yifeng Huang, Ruofan Jiang +1
We study the cohomology of , the moduli space of commuting -by- matrices satisfying the equations defining a quasi-projective scheme . This space can be viewed as…
On the mechanical creation of mathematical concepts
Asvin G
Any act of problem-solving combines prior knowledge, local search, and a third element that is less often discussed: the extraction of information from search to update understandi…
Computational Platonism
Asvin G
We offer a view of mathematics as an experimental science where axioms play the role of foundational theories like general relativity and quantum mechanics in physics. Under this v…