Research map / Label-efficient and robust learning / Self-supervised and contrastive
Enhancing Contrastive Learning with Variable Similarity
Haowen Cui, Shuo Chen, Jun Li, Jian Yang
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes variable-similarity contrastive learning that adjusts positive-pair similarity targets according to the augmentation strength, rather than treating all augmented views of an instance as equally similar.
Problem
Standard contrastive learning assumes all augmented views of the same instance remain semantically invariant, but augmentations of varying extent can introduce semantic discrepancies or even conflicts.
Approach
Treats positive pairs as having variable (graded) similarity depending on augmentation extent instead of uniform positive similarity, so that strongly augmented views are not forced to match fully.
Closest papers by meaning
- What Should Not Be Contrastive in Contrastive Learning (ICLR 2021 · similarity 0.90)
- Contrastive Learning with Stronger Augmentations (TPAMI 2022 · similarity 0.89)
- A Statistical Theory of Contrastive Learning via Approximate Sufficient Statistics (NeurIPS 2025 · similarity 0.89)
- Variational Supervised Contrastive Learning (NeurIPS 2025 · similarity 0.88)
- On the Similarities of Embeddings in Contrastive Learning (ICML 2025 · similarity 0.88)
- An Augmentation-Aware Theory for Self-Supervised Contrastive Learning (ICML 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
