# Ladder Loss for Coherent Visual-Semantic Embedding

arXiv: https://arxiv.org/abs/1911.07528 · HTML: https://mlsanity.com/paper/1911.07528/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Nov 2019.

## Built on by (most cited first, 15 of 15)

- [CoNT: Contrastive Neural Text Generation](https://mlsanity.com/paper/2205.14690/) (2022)
- [Is An Image Worth Five Sentences? A New Look into Semantics for Image-Text Matching](https://mlsanity.com/paper/2110.02623/) (2021)
- [Enhancing Adversarial Robustness for Deep Metric Learning](https://mlsanity.com/paper/2203.01439/) (2022)
- [Practical Relative Order Attack in Deep Ranking](https://mlsanity.com/paper/2103.05248/) (2021)
- [Adversarial Attack and Defense in Deep Ranking](https://mlsanity.com/paper/2106.03614/) (2021)
- [Domain Adaptation in Multi-View Embedding for Cross-Modal Video Retrieval](https://mlsanity.com/paper/2110.12812/) (2021)
- [SwAMP: Swapped Assignment of Multi-Modal Pairs for Cross-Modal Retrieval](https://mlsanity.com/paper/2111.05814/) (2021)
- [Deep Boosting Learning: A Brand-new Cooperative Approach for Image-Text Matching](https://mlsanity.com/paper/2404.18114/) (2024)
- [Composed Multi-modal Retrieval: A Survey of Approaches and Applications](https://mlsanity.com/paper/2503.01334/) (2025)
- [Intra-Modal Constraint Loss For Image-Text Retrieval](https://mlsanity.com/paper/2207.05024/) (2022)
- [Mathematical Justification of Hard Negative Mining via Isometric Approximation Theorem](https://mlsanity.com/paper/2210.11173/) (2022)
- [Image-Text Retrieval with Binary and Continuous Label Supervision](https://mlsanity.com/paper/2210.11319/) (2022)
- [An embedding for EEG signals learned using a triplet loss](https://mlsanity.com/paper/2304.06495/) (2023)
- [Reranking Overgenerated Responses for End-to-End Task-Oriented Dialogue Systems](https://mlsanity.com/paper/2211.03648/) (2022)
- [Descriptive Image-Text Matching with Graded Contextual Similarity](https://mlsanity.com/paper/2505.09997/) (2025)

## Abstract (arXiv)

For visual-semantic embedding, the existing methods normally treat the relevance between queries and candidates in a bipolar way -- relevant or irrelevant, and all "irrelevant" candidates are uniformly pushed away from the query by an equal margin in the embedding space, regardless of their various proximity to the query. This practice disregards relatively discriminative information and could lead to suboptimal ranking in the retrieval results and poorer user experience, especially in the long-tail query scenario where a matching candidate may not necessarily exist. In this paper, we introduce a continuous variable to model the relevance degree between queries and multiple candidates, and propose to learn a coherent embedding space, where candidates with higher relevance degrees are mapped closer to the query than those with lower relevance degrees. In particular, the new ladder loss is proposed by extending the triplet loss inequality to a more general inequality chain, which implements variable push-away margins according to respective relevance degrees. In addition, a proper Coherent Score metric is proposed to better measure the ranking results including those "irrelevant" candidates. Extensive experiments on multiple datasets validate the efficacy of our proposed method, which achieves significant improvement over existing state-of-the-art methods.