Pith. sign in

Vlmo: Unified vision-language pre-training with mixture-of-modality- experts.Advances in Neural Information Processing Systems, 35:32897– 32912, 2022

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

citation-role summary

background 1

citation-polarity summary

fields

cs.CV 1

years

2025 1

verdicts

REJECT 1

roles

background 1

polarities

unclear 1

representative citing papers

Dynamic Double Space Tower

cs.CV · 2025-06-13 · reject · novelty 3.0

The paper claims a four-layer Gestalt-based tower can replace attention in VQA and lift a 3B model to state-of-the-art spatial reasoning, but provides no reproducible method or consistent results.

citing papers explorer

Showing 1 of 1 citing paper.

  • Dynamic Double Space Tower cs.CV · 2025-06-13 · reject · none · ref 1

    The paper claims a four-layer Gestalt-based tower can replace attention in VQA and lift a 3B model to state-of-the-art spatial reasoning, but provides no reproducible method or consistent results.