pith. sign in

International Conference on Machine Learning , pages=

3 Pith papers cite this work. Polarity classification is still indexing.

3 Pith papers citing it

citation-role summary

background 1

citation-polarity summary

fields

cs.LG 3

years

2026 3

verdicts

UNVERDICTED 3

roles

background 1

polarities

background 1

representative citing papers

A Unified Geometric Framework for Weighted Contrastive Learning

cs.LG · 2026-05-13 · unverdicted · novelty 8.0

Weighted InfoNCE objectives realize specific target geometries in embedding space, with SupCon producing size-dependent inter-class similarities under imbalance while Soft SupCon and certain continuous variants preserve regular simplex or unique optima.

citing papers explorer

Showing 3 of 3 citing papers.

  • A Unified Geometric Framework for Weighted Contrastive Learning cs.LG · 2026-05-13 · unverdicted · none · ref 1

    Weighted InfoNCE objectives realize specific target geometries in embedding space, with SupCon producing size-dependent inter-class similarities under imbalance while Soft SupCon and certain continuous variants preserve regular simplex or unique optima.

  • Optimal Representations for Generalized Contrastive Learning with Imbalanced Datasets cs.LG · 2026-05-11 · unverdicted · none · ref 5

    In generalized contrastive learning with imbalanced classes, optimal representations collapse to class means whose angular geometry is determined by class proportions via convex optimization, and extreme imbalance causes all minority classes to collapse to one vector.

  • Neural Collapse by Design: Learning Class Prototypes on the Hypersphere cs.LG · 2026-05-19 · unverdicted · none · ref 42 · 2 links

    Supervised classification reaches neural collapse by design via normalized prototype losses on the hypersphere, outperforming CE and SCL on ImageNet-1K and other benchmarks with faster convergence and better transfer.