Proto-Interpretation: The Temporality of Large Language Model Inference
ACM AI Letters · 31.03.2026 · 1 (1) · 1-8
DOI: 10.1145/3789666 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM AI Letters, online erschienen am 26.03.2026. Verfasst von Mattias Rost. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 11 Literaturangaben.
Abstract (Originalsprache)
We show that autoregressive generation in large language models exhibits a temporal structure: each token is not only conditioned on the past but also reshapes the future continuation space. We call this process proto-interpretation : the probabilistic redistribution across competing continuations through which the model gradually commits to one emerging branch of meaning. Using a minimal ambiguity case, we demonstrate branch competition and sequential commitment during inference. These findings reveal meaning in LLMs as a dynamic, temporally unfolding process, shifting interpretability from static model states to inference-time dynamics.
Zitieren
APA
Rost, M. (2026). Proto-Interpretation: The Temporality of Large Language Model Inference. ACM AI Letters, 1(1), 1-8. https://doi.org/10.1145/3789666
Vancouver
Rost M. Proto-Interpretation: The Temporality of Large Language Model Inference. ACM AI Letters. 2026;1(1):1-8. doi:10.1145/3789666
BibTeX
@article{Rost2026,
author = {Rost, Mattias},
title = {Proto-Interpretation: The Temporality of Large Language Model Inference},
journal = {ACM AI Letters},
year = {2026},
volume = {1},
number = {1},
pages = {1--8},
doi = {10.1145/3789666}
}
EUVIMED