Research map / Label-efficient and robust learning / Recognition and long-tail
Conditional Similarity Networks
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use conditional semantic masks instead of a single shared triplet space | Triplet embedding network · Zappos50k test triplets | Error Rate: 23.72% -> 10.73% | Helped |
| Train separate specialist triplet networks for each similarity notion | Standard Triplet Network · Zappos50k test triplets | Error Rate: 23.72% -> 11.35% | Helped |
| Use fixed disjoint masks rather than a single unconditioned triplet space | Standard Triplet Network · Zappos50k test triplets | Error Rate: 23.72% -> 10.79% | Helped |
| Learn masks jointly instead of fixing them as disjoint (ablation) | CSN fixed disjoint masks · Zappos50k test triplets | Error Rate: 10.79% -> 10.73% | Helped |
| Fine-tune with a standard triplet loss across contradictory similarity notions | ImageNet-pretrained ResNet features · Zappos50k brand classification | Top 1 Accuracy: 54.00% -> 49.08% | Hurt |
| Fine-tune with conditional similarity masks rather than standard triplet training | ImageNet-pretrained ResNet features · Zappos50k brand classification | Top 1 Accuracy: 49.08% -> 53.67% | Helped |
Closest papers by meaning
- Decoupling Semantic Similarity from Spatial Alignment for Neural Networks. (NeurIPS 2024 · similarity 0.87)
- Representational Similarity via Interpretable Visual Concepts (ICLR 2025 · similarity 0.87)
- $\mathbb{X}$-Sample Contrastive Loss: Improving Contrastive Learning with Sample Similarity Graphs (ICLR 2025 · similarity 0.86)
- Interpretable Measures of Conceptual Similarity by Complexity-Constrained Descriptive Auto-Encoding (CVPR 2024 · similarity 0.86)
- On the Similarities of Embeddings in Contrastive Learning (ICML 2025 · similarity 0.85)
- Going Beyond Neural Network Feature Similarity: The Network Feature Complexity and Its Interpretation Using Category Theory (ICLR 2024 · similarity 0.85)
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.