New York, New York
New York University
NYU is where the field gets audited: benchmarks that expose what models cannot do, and the convolutional groundwork that started the modern run.
Key findings
- 2018
GLUE and SuperGLUE
Wang, Singh, Bowman et al.
A multi-task benchmark suite for language understanding, replaced within two years by a harder version because models saturated it.
Why it matters here · The saturation cycle we now watch quarter by quarter in the change feed.
Human baseline passed in ~18 months - 1989
Convolutional networks and backpropagation for digits
Yann LeCun and collaborators
Trained a convolutional network end-to-end on handwritten digits, deployed to read bank cheques at scale.
Why it matters here · The first commercially deployed deep network — the proof the whole industry was built on.
- 2015
SNLI — inference at dataset scale
Bowman, Angeli, Potts, Manning
570,000 human-written sentence pairs labelled for entailment, contradiction or neutrality.
Why it matters here · Made reasoning measurable before models could reason; still a probe for hallucination behaviour.
570k labelled pairs
What to follow
How to cite this page
Free to cite and reuse under CC BY 4.0. Permalink: https://tomorrow.aliensquad.ai/academia/nyu
Tomorrow. (2026). New York University — key findings and research feeds [Research tracker entry]. AlienSquad. Retrieved 2026-09-13, from https://tomorrow.aliensquad.ai/academia/nyu
@misc{tomorrow-academia-nyu,
author = {{Tomorrow}},
title = {New York University — key findings and research feeds},
year = {2026},
publisher = {AlienSquad},
howpublished = {\url{https://tomorrow.aliensquad.ai/academia/nyu}},
note = {Accessed: 2026-09-13}
}