# RE-TRIANGLE: Does TRIANGLE Enable Multimodal Alignment Beyond Cosine Similarity in Retrieval?

arXiv: https://arxiv.org/abs/2605.27436 · HTML: https://mlsanity.com/paper/2605.27436/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 15 of its 25 references that are in the map.
- First posted on arXiv 22 May 2026.

## Builds on (most cited first, 15 of 15)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://mlsanity.com/paper/2102.05918/) (2021)
- [Localizing Moments in Video with Natural Language](https://mlsanity.com/paper/1708.01641/) (2017)
- [CLIP4Clip: An Empirical Study of CLIP for End to End Video Clip Retrieval](https://mlsanity.com/paper/2104.08860/) (2021)
- [Towards Automatic Learning of Procedures from Web Instructional Videos](https://mlsanity.com/paper/1703.09788/) (2017)
- [PointCLIP: Point Cloud Understanding by CLIP](https://mlsanity.com/paper/2112.02413/) (2021)
- [VATEX: A Large-Scale, High-Quality Multilingual Dataset for Video-and-Language Research](https://mlsanity.com/paper/1904.03493/) (2019)
- [BEATs: Audio Pre-Training with Acoustic Tokenizers](https://mlsanity.com/paper/2212.09058/) (2022)
- [VAST: A Vision-Audio-Subtitle-Text Omni-Modality Foundation Model and Dataset](https://mlsanity.com/paper/2305.18500/) (2023)
- [Unifying Multimodal Retrieval via Document Screenshot Embedding](https://mlsanity.com/paper/2406.11251/) (2024)
- [Gramian Multimodal Representation Learning and Alignment](https://mlsanity.com/paper/2412.11959/) (2024)
- [Contrasting with Symile: Simple Model-Agnostic Representation Learning for Unlimited Modalities](https://mlsanity.com/paper/2411.01053/) (2024)
- [A TRIANGLE Enables Multimodal Alignment Beyond Cosine Similarity](https://mlsanity.com/paper/2509.24734/) (2025)
- [Weighted Point Set Embedding for Multimodal Contrastive Learning Toward Optimal Similarity Metric](https://mlsanity.com/paper/2404.19228/) (2024)
- [Reproducibility, Replicability, and Insights into Visual Document Retrieval with Late Interaction](https://mlsanity.com/paper/2505.07730/) (2025)

## Abstract (arXiv)

Multimodal alignment is critical for bridging the semantic gap in information retrieval. However, traditional pairwise strategies introduce a geometric blind spot: while they align anchor modalities (e.g., text) with others, they lack constraints to enforce mutual consistency between peripheral modalities (e.g., video and audio). The TRIANGLE framework addresses this by minimizing the area of modality triplets on a hypersphere to enforce holistic alignment. In this reproducibility study, we verify the robustness of this geometric objective for retrieval tasks. We confirm that TRIANGLE outperforms pairwise baselines in zero-shot settings, achieving Recall@1 gains of up to +8.7 points, though benefits are domain-dependent. However, we fail to reproduce the reported learning-from-scratch results. Analysis using a synthetic toy dataset attributes this to instability when jointly optimizing geometric alignment with Data-Text Matching (DTM) loss. Furthermore, we find that cosine regularization primarily stabilizes text-to-video retrieval, and fine-tuning with domain supervision amplifies geometric benefits but reduces cross-dataset generalization. Our findings support the efficacy of geometric alignment while highlighting critical optimization sensitivities. Code available at https://github.com/ARIJIT00171/RE-TRIANGLE.